diff --git "a/model.safetensors.index.json" "b/model.safetensors.index.json"
new file mode 100644--- /dev/null
+++ "b/model.safetensors.index.json"
@@ -0,0 +1,5765 @@
+{
+  "metadata": {
+    "total_size": 451415080160
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00075.safetensors",
+    "model.embed_tokens.scales": "model-00001-of-00075.safetensors",
+    "model.embed_tokens.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.up_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.up_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.down_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.down_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.gate_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp.gate_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.g_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.g_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.g_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.dt_bias": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_b_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_b_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_b_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_a_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_a_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.f_a_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.A_log": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.o_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.o_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.o_norm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.b_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.b_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.b_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.v_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.v_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.k_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.k_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.q_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.q_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attention_res.proj_weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attention_res.norm_weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp_res.proj_weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.mlp_res.norm_weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.q_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.k_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.0.self_attn.v_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_norm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_up_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_up_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_up_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_down_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_down_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.routed_expert_down_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.up_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.up_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.up_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.down_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.down_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.down_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.gate_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.gate_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.shared_experts.gate_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.gate.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.gate.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.g_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.g_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.g_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.dt_bias": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_b_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_b_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_b_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_a_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_a_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.f_a_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.A_log": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.o_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.o_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.o_norm.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.b_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.b_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.b_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.v_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.v_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.k_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.k_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.q_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.q_proj.biases": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attention_res.proj_weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attention_res.norm_weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.mlp_res.proj_weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.mlp_res.norm_weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.e_score_correction_bias": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.q_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.k_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.self_attn.v_conv.conv.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.up_proj.weight": "model-00001-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.up_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.down_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.1.block_sparse_moe.switch_mlp.down_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_norm.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_up_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_up_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_up_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_down_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_down_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.routed_expert_down_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.up_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.up_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.up_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.down_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.down_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.down_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.gate_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.gate_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.shared_experts.gate_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.gate.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.gate.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.gate.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.g_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.g_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.g_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.dt_bias": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_b_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_b_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_b_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_a_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_a_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.f_a_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.A_log": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.o_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.o_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.o_norm.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.b_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.b_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.b_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.v_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.v_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.k_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.k_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.q_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.q_proj.biases": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attention_res.proj_weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attention_res.norm_weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.mlp_res.proj_weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.mlp_res.norm_weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.e_score_correction_bias": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.q_conv.conv.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.k_conv.conv.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.self_attn.v_conv.conv.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.up_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.up_proj.scales": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.down_proj.weight": "model-00002-of-00075.safetensors",
+    "model.layers.2.block_sparse_moe.switch_mlp.down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.kv_a_proj_with_mqa.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.kv_a_proj_with_mqa.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.kv_a_proj_with_mqa.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_b_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_b_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_b_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_norm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_up_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_up_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_up_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_down_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.routed_expert_down_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.up_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.up_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.up_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.down_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.down_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.gate_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.gate_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.shared_experts.gate_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.gate.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.gate.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.gate.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.g_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.g_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.g_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.o_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.o_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.kv_a_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_a_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_a_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_a_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.q_a_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attention_res.proj_weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attention_res.norm_weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.mlp_res.proj_weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.mlp_res.norm_weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.e_score_correction_bias": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.embed_q.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.embed_q.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.embed_q.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.unembed_out.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.unembed_out.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.self_attn.unembed_out.biases": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.up_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.up_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.down_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.3.block_sparse_moe.switch_mlp.down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_norm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_up_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_up_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_up_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_down_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.routed_expert_down_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.up_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.up_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.up_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.down_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.down_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.down_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.gate_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.gate_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.shared_experts.gate_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.gate.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.gate.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.gate.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.g_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.g_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.g_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.dt_bias": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_b_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_b_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_b_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_a_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_a_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.f_a_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.A_log": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.o_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.o_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.o_norm.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.b_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.b_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.b_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.v_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.v_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.k_proj.scales": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.k_proj.biases": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00003-of-00075.safetensors",
+    "model.layers.4.self_attn.q_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attn.q_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attention_res.proj_weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attention_res.norm_weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.mlp_res.proj_weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.mlp_res.norm_weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.e_score_correction_bias": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attn.q_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attn.k_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.self_attn.v_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.up_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.up_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.down_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.4.block_sparse_moe.switch_mlp.down_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_norm.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_up_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_up_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_up_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_down_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_down_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.routed_expert_down_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.up_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.up_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.up_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.down_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.down_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.down_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.gate_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.gate_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.shared_experts.gate_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.gate.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.gate.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.gate.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.g_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.g_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.g_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.dt_bias": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_b_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_b_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_b_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_a_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_a_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.f_a_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.A_log": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.o_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.o_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.o_norm.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.b_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.b_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.b_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.v_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.v_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.k_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.k_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.q_proj.scales": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.q_proj.biases": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attention_res.proj_weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attention_res.norm_weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.mlp_res.proj_weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.mlp_res.norm_weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.e_score_correction_bias": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.q_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.k_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.self_attn.v_conv.conv.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00004-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.up_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.up_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.down_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.5.block_sparse_moe.switch_mlp.down_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_norm.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_up_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_up_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_up_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_down_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_down_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.routed_expert_down_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.up_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.up_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.up_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.down_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.down_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.down_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.gate_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.gate_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.shared_experts.gate_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.gate.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.gate.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.gate.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.g_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.g_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.g_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.dt_bias": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_b_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_b_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_b_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_a_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_a_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.f_a_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.A_log": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.o_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.o_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.o_norm.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.b_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.b_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.b_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.v_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.v_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.k_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.k_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.q_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.q_proj.biases": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attention_res.proj_weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attention_res.norm_weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.mlp_res.proj_weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.mlp_res.norm_weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.e_score_correction_bias": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.q_conv.conv.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.k_conv.conv.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.self_attn.v_conv.conv.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.up_proj.weight": "model-00005-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.up_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.down_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.6.block_sparse_moe.switch_mlp.down_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.kv_a_proj_with_mqa.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.kv_a_proj_with_mqa.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.kv_a_proj_with_mqa.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_b_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_b_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_b_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_norm.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_up_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_up_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_up_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_down_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_down_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.routed_expert_down_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.up_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.up_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.up_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.down_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.down_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.down_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.gate_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.gate_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.shared_experts.gate_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.gate.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.gate.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.gate.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.g_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.g_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.g_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.o_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.o_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.kv_a_layernorm.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_a_layernorm.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_a_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_a_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.q_a_proj.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attention_res.proj_weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attention_res.norm_weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.mlp_res.proj_weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.mlp_res.norm_weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.e_score_correction_bias": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.embed_q.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.embed_q.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.embed_q.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.unembed_out.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.unembed_out.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.self_attn.unembed_out.biases": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.up_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.up_proj.scales": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.down_proj.weight": "model-00006-of-00075.safetensors",
+    "model.layers.7.block_sparse_moe.switch_mlp.down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_norm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_up_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_up_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_up_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_down_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.routed_expert_down_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.up_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.up_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.up_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.down_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.down_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.gate_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.gate_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.shared_experts.gate_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.gate.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.gate.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.gate.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.g_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.g_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.g_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.dt_bias": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_b_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_b_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_b_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_a_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_a_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.f_a_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.A_log": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.o_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.o_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.o_norm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.b_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.b_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.b_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.v_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.v_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.k_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.k_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.q_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.q_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attention_res.proj_weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attention_res.norm_weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.mlp_res.proj_weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.mlp_res.norm_weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.e_score_correction_bias": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.q_conv.conv.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.k_conv.conv.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.self_attn.v_conv.conv.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.up_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.up_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.down_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.8.block_sparse_moe.switch_mlp.down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_norm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_up_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_up_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_up_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_down_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.routed_expert_down_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.up_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.up_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.up_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.down_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.down_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.down_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.gate_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.gate_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.shared_experts.gate_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.gate.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.gate.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.gate.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.g_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.g_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.g_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.dt_bias": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_b_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_b_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_b_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_a_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_a_proj.scales": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.f_a_proj.biases": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.A_log": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00007-of-00075.safetensors",
+    "model.layers.9.self_attn.o_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.o_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.o_norm.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.b_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.b_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.b_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.v_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.v_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.k_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.k_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.q_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.q_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attention_res.proj_weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attention_res.norm_weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.mlp_res.proj_weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.mlp_res.norm_weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.e_score_correction_bias": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.q_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.k_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.self_attn.v_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.up_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.up_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.down_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.9.block_sparse_moe.switch_mlp.down_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_norm.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_up_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_up_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_up_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_down_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_down_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.routed_expert_down_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.up_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.up_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.up_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.down_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.down_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.down_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.gate_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.gate_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.shared_experts.gate_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.gate.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.gate.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.gate.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.g_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.g_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.g_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.dt_bias": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_b_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_b_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_b_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_a_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_a_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.f_a_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.A_log": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.o_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.o_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.o_norm.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.b_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.b_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.b_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.v_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.v_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.k_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.k_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.q_proj.scales": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.q_proj.biases": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attention_res.proj_weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attention_res.norm_weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.mlp_res.proj_weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.mlp_res.norm_weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.e_score_correction_bias": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.q_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.k_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.self_attn.v_conv.conv.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00008-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.up_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.up_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.down_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.10.block_sparse_moe.switch_mlp.down_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.kv_a_proj_with_mqa.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.kv_a_proj_with_mqa.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.kv_a_proj_with_mqa.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_b_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_b_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_b_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_norm.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_up_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_up_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_up_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_down_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_down_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.routed_expert_down_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.up_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.up_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.up_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.down_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.down_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.down_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.gate_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.gate_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.shared_experts.gate_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.gate.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.gate.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.gate.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.g_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.g_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.g_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.o_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.o_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.kv_a_layernorm.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_a_layernorm.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_a_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_a_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.q_a_proj.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attention_res.proj_weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attention_res.norm_weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.mlp_res.proj_weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.mlp_res.norm_weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.e_score_correction_bias": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.embed_q.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.embed_q.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.embed_q.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.unembed_out.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.unembed_out.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.self_attn.unembed_out.biases": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.up_proj.weight": "model-00009-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.up_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.down_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.11.block_sparse_moe.switch_mlp.down_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_norm.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_up_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_up_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_up_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_down_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_down_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.routed_expert_down_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.up_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.up_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.up_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.down_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.down_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.down_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.gate_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.gate_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.shared_experts.gate_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.gate.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.gate.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.gate.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.g_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.g_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.g_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.dt_bias": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_b_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_b_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_b_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_a_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_a_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.f_a_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.A_log": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.o_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.o_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.o_norm.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.b_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.b_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.b_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.v_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.v_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.k_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.k_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.q_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.q_proj.biases": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attention_res.proj_weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attention_res.norm_weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.mlp_res.proj_weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.mlp_res.norm_weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.e_score_correction_bias": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.q_conv.conv.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.k_conv.conv.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.self_attn.v_conv.conv.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.up_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.up_proj.scales": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.down_proj.weight": "model-00010-of-00075.safetensors",
+    "model.layers.12.block_sparse_moe.switch_mlp.down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_norm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_up_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_up_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_up_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_down_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.routed_expert_down_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.up_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.up_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.up_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.down_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.down_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.gate_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.gate_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.shared_experts.gate_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.gate.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.gate.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.gate.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.g_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.g_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.g_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.dt_bias": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_b_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_b_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_b_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_a_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_a_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.f_a_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.A_log": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.o_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.o_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.o_norm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.b_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.b_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.b_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.v_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.v_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.k_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.k_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.q_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.q_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attention_res.proj_weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attention_res.norm_weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.mlp_res.proj_weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.mlp_res.norm_weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.e_score_correction_bias": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.q_conv.conv.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.k_conv.conv.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.self_attn.v_conv.conv.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.up_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.up_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.down_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.13.block_sparse_moe.switch_mlp.down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_norm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_up_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_up_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_up_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_down_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.routed_expert_down_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.up_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.up_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.up_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.down_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.down_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.down_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.gate_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.gate_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.shared_experts.gate_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.gate.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.gate.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.gate.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.g_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.g_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.g_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.dt_bias": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_b_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_b_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_b_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_a_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_a_proj.scales": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.f_a_proj.biases": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.A_log": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00011-of-00075.safetensors",
+    "model.layers.14.self_attn.o_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.o_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.o_norm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.b_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.b_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.b_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.v_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.v_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.k_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.k_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.q_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.q_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attention_res.proj_weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attention_res.norm_weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.mlp_res.proj_weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.mlp_res.norm_weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.e_score_correction_bias": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.q_conv.conv.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.k_conv.conv.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.self_attn.v_conv.conv.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.up_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.up_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.down_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.14.block_sparse_moe.switch_mlp.down_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.kv_a_proj_with_mqa.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.kv_a_proj_with_mqa.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.kv_a_proj_with_mqa.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_b_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_b_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_b_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_norm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_up_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_up_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_up_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_down_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_down_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.routed_expert_down_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.up_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.up_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.up_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.down_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.down_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.down_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.gate_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.gate_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.shared_experts.gate_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.gate.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.gate.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.gate.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.g_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.g_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.g_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.o_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.o_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.kv_a_layernorm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_a_layernorm.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_a_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_a_proj.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.q_a_proj.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attention_res.proj_weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attention_res.norm_weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.mlp_res.proj_weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.mlp_res.norm_weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.e_score_correction_bias": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.embed_q.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.embed_q.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.embed_q.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.unembed_out.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.unembed_out.scales": "model-00012-of-00075.safetensors",
+    "model.layers.15.self_attn.unembed_out.biases": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00012-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.up_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.up_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.down_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.15.block_sparse_moe.switch_mlp.down_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_norm.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_up_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_up_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_up_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_down_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_down_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.routed_expert_down_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.up_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.up_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.up_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.down_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.down_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.down_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.gate_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.gate_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.shared_experts.gate_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.gate.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.gate.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.gate.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.g_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.g_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.g_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.dt_bias": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_b_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_b_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_b_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_a_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_a_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.f_a_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.A_log": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.o_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.o_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.o_norm.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.b_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.b_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.b_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.v_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.v_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.k_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.k_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.q_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.q_proj.biases": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attention_res.proj_weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attention_res.norm_weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.mlp_res.proj_weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.mlp_res.norm_weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.e_score_correction_bias": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.q_conv.conv.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.k_conv.conv.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.self_attn.v_conv.conv.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.up_proj.weight": "model-00013-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.up_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.down_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.16.block_sparse_moe.switch_mlp.down_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_norm.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_up_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_up_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_up_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_down_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_down_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.routed_expert_down_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.up_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.up_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.up_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.down_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.down_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.down_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.gate_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.gate_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.shared_experts.gate_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.gate.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.gate.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.gate.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.g_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.g_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.g_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.dt_bias": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_b_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_b_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_b_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_a_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_a_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.f_a_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.A_log": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.o_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.o_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.o_norm.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.b_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.b_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.b_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.v_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.v_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.k_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.k_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.q_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.q_proj.biases": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attention_res.proj_weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attention_res.norm_weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.mlp_res.proj_weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.mlp_res.norm_weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.e_score_correction_bias": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.q_conv.conv.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.k_conv.conv.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.self_attn.v_conv.conv.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.up_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.up_proj.scales": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.down_proj.weight": "model-00014-of-00075.safetensors",
+    "model.layers.17.block_sparse_moe.switch_mlp.down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_norm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_up_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_up_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_up_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_down_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.routed_expert_down_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.up_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.up_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.up_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.down_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.down_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.gate_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.gate_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.shared_experts.gate_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.gate.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.gate.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.gate.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.g_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.g_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.g_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.dt_bias": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_b_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_b_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_b_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_a_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_a_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.f_a_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.A_log": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.o_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.o_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.o_norm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.b_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.b_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.b_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.v_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.v_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.k_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.k_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.q_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.q_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attention_res.proj_weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attention_res.norm_weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.mlp_res.proj_weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.mlp_res.norm_weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.e_score_correction_bias": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.q_conv.conv.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.k_conv.conv.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.self_attn.v_conv.conv.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.up_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.up_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.down_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.18.block_sparse_moe.switch_mlp.down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.kv_a_proj_with_mqa.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.kv_a_proj_with_mqa.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.kv_a_proj_with_mqa.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.q_b_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.q_b_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.q_b_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_norm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_up_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_up_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_up_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_down_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.routed_expert_down_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.up_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.up_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.up_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.down_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.down_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.down_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.gate_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.gate_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.shared_experts.gate_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.gate.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.gate.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.gate.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.g_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.g_proj.scales": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.g_proj.biases": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00015-of-00075.safetensors",
+    "model.layers.19.self_attn.o_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.o_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.kv_a_layernorm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.q_a_layernorm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.q_a_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.q_a_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.q_a_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attention_res.proj_weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attention_res.norm_weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.mlp_res.proj_weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.mlp_res.norm_weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.e_score_correction_bias": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.embed_q.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.embed_q.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.embed_q.biases": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.unembed_out.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.unembed_out.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.self_attn.unembed_out.biases": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.up_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.up_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.down_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.19.block_sparse_moe.switch_mlp.down_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_norm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_up_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_up_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_up_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_down_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_down_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.routed_expert_down_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.up_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.up_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.up_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.down_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.down_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.down_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.gate_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.gate_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.shared_experts.gate_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.gate.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.gate.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.gate.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.g_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.g_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.g_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.dt_bias": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_b_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_b_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_b_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_a_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_a_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.f_a_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.A_log": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.o_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.o_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.o_norm.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.b_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.b_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.b_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.v_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.v_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.k_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.k_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.q_proj.scales": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.q_proj.biases": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attention_res.proj_weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attention_res.norm_weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.mlp_res.proj_weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.mlp_res.norm_weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.e_score_correction_bias": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.q_conv.conv.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.k_conv.conv.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.self_attn.v_conv.conv.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00016-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.up_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.up_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.down_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.20.block_sparse_moe.switch_mlp.down_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_norm.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_up_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_up_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_up_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_down_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_down_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.routed_expert_down_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.up_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.up_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.up_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.down_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.down_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.down_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.gate_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.gate_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.shared_experts.gate_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.gate.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.gate.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.gate.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.g_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.g_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.g_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.dt_bias": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_b_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_b_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_b_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_a_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_a_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.f_a_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.A_log": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.o_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.o_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.o_norm.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.b_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.b_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.b_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.v_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.v_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.k_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.k_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.q_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.q_proj.biases": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attention_res.proj_weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attention_res.norm_weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.mlp_res.proj_weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.mlp_res.norm_weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.e_score_correction_bias": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.q_conv.conv.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.k_conv.conv.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.self_attn.v_conv.conv.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.up_proj.weight": "model-00017-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.up_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.down_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.21.block_sparse_moe.switch_mlp.down_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_norm.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_up_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_up_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_up_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_down_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_down_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.routed_expert_down_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.up_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.up_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.up_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.down_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.down_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.down_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.gate_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.gate_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.shared_experts.gate_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.gate.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.gate.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.gate.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.g_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.g_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.g_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.dt_bias": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_b_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_b_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_b_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_a_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_a_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.f_a_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.A_log": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.o_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.o_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.o_norm.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.b_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.b_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.b_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.v_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.v_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.k_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.k_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.q_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.q_proj.biases": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attention_res.proj_weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attention_res.norm_weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.mlp_res.proj_weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.mlp_res.norm_weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.e_score_correction_bias": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.q_conv.conv.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.k_conv.conv.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.self_attn.v_conv.conv.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.up_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.up_proj.scales": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.down_proj.weight": "model-00018-of-00075.safetensors",
+    "model.layers.22.block_sparse_moe.switch_mlp.down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.kv_a_proj_with_mqa.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.kv_a_proj_with_mqa.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.kv_a_proj_with_mqa.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_b_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_b_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_b_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_norm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_up_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_up_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_up_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_down_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.routed_expert_down_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.up_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.up_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.up_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.down_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.down_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.gate_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.gate_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.shared_experts.gate_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.gate.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.gate.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.gate.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.g_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.g_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.g_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.o_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.o_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.kv_a_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_a_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_a_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_a_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.q_a_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attention_res.proj_weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attention_res.norm_weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.mlp_res.proj_weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.mlp_res.norm_weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.e_score_correction_bias": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.embed_q.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.embed_q.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.embed_q.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.unembed_out.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.unembed_out.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.self_attn.unembed_out.biases": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.up_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.up_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.down_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.23.block_sparse_moe.switch_mlp.down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_norm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_up_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_up_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_up_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_down_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.routed_expert_down_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.up_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.up_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.up_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.down_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.down_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.down_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.gate_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.gate_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.shared_experts.gate_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.gate.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.gate.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.gate.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.g_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.g_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.g_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.dt_bias": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_b_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_b_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_b_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_a_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_a_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.f_a_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.A_log": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.o_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.o_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.o_norm.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.b_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.b_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.b_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.v_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.v_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.k_proj.scales": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.k_proj.biases": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00019-of-00075.safetensors",
+    "model.layers.24.self_attn.q_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attn.q_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attention_res.proj_weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attention_res.norm_weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.mlp_res.proj_weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.mlp_res.norm_weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.e_score_correction_bias": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attn.q_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attn.k_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.self_attn.v_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.up_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.up_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.down_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.24.block_sparse_moe.switch_mlp.down_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_norm.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_up_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_up_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_up_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_down_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_down_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.routed_expert_down_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.up_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.up_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.up_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.down_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.down_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.down_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.gate_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.gate_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.shared_experts.gate_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.gate.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.gate.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.gate.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.g_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.g_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.g_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.dt_bias": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_b_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_b_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_b_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_a_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_a_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.f_a_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.A_log": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.o_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.o_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.o_norm.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.b_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.b_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.b_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.v_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.v_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.k_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.k_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.q_proj.scales": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.q_proj.biases": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attention_res.proj_weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attention_res.norm_weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.mlp_res.proj_weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.mlp_res.norm_weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.e_score_correction_bias": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.q_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.k_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.self_attn.v_conv.conv.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00020-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.up_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.up_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.down_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.25.block_sparse_moe.switch_mlp.down_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_norm.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_up_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_up_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_up_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_down_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_down_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.routed_expert_down_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.up_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.up_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.up_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.down_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.down_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.down_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.gate_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.gate_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.shared_experts.gate_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.gate.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.gate.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.gate.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.g_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.g_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.g_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.dt_bias": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_b_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_b_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_b_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_a_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_a_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.f_a_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.A_log": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.o_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.o_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.o_norm.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.b_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.b_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.b_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.v_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.v_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.k_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.k_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.q_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.q_proj.biases": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attention_res.proj_weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attention_res.norm_weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.mlp_res.proj_weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.mlp_res.norm_weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.e_score_correction_bias": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.q_conv.conv.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.k_conv.conv.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.self_attn.v_conv.conv.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.up_proj.weight": "model-00021-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.up_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.down_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.26.block_sparse_moe.switch_mlp.down_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.kv_a_proj_with_mqa.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.kv_a_proj_with_mqa.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.kv_a_proj_with_mqa.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_b_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_b_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_b_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_norm.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_up_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_up_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_up_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_down_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_down_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.routed_expert_down_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.up_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.up_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.up_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.down_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.down_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.down_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.gate_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.gate_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.shared_experts.gate_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.gate.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.gate.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.gate.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.g_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.g_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.g_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.o_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.o_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.kv_a_layernorm.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_a_layernorm.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_a_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_a_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.q_a_proj.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attention_res.proj_weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attention_res.norm_weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.mlp_res.proj_weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.mlp_res.norm_weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.e_score_correction_bias": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.embed_q.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.embed_q.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.embed_q.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.unembed_out.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.unembed_out.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.self_attn.unembed_out.biases": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.up_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.up_proj.scales": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.down_proj.weight": "model-00022-of-00075.safetensors",
+    "model.layers.27.block_sparse_moe.switch_mlp.down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_norm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_up_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_up_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_up_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_down_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.routed_expert_down_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.up_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.up_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.up_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.down_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.down_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.gate_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.gate_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.shared_experts.gate_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.gate.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.gate.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.gate.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.g_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.g_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.g_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.dt_bias": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_b_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_b_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_b_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_a_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_a_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.f_a_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.A_log": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.o_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.o_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.o_norm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.b_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.b_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.b_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.v_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.v_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.k_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.k_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.q_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.q_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attention_res.proj_weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attention_res.norm_weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.mlp_res.proj_weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.mlp_res.norm_weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.e_score_correction_bias": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.q_conv.conv.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.k_conv.conv.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.self_attn.v_conv.conv.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.up_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.up_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.down_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.28.block_sparse_moe.switch_mlp.down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_norm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_up_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_up_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_up_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_down_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.routed_expert_down_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.up_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.up_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.up_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.down_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.down_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.down_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.gate_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.gate_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.shared_experts.gate_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.gate.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.gate.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.gate.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.g_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.g_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.g_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.dt_bias": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_b_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_b_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_b_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_a_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_a_proj.scales": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.f_a_proj.biases": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.A_log": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00023-of-00075.safetensors",
+    "model.layers.29.self_attn.o_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.o_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.o_norm.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.b_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.b_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.b_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.v_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.v_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.k_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.k_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.q_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.q_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attention_res.proj_weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attention_res.norm_weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.mlp_res.proj_weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.mlp_res.norm_weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.e_score_correction_bias": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.q_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.k_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.self_attn.v_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.up_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.up_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.down_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.29.block_sparse_moe.switch_mlp.down_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_norm.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_up_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_up_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_up_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_down_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_down_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.routed_expert_down_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.up_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.up_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.up_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.down_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.down_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.down_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.gate_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.gate_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.shared_experts.gate_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.gate.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.gate.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.gate.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.g_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.g_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.g_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.dt_bias": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_b_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_b_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_b_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_a_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_a_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.f_a_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.A_log": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.o_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.o_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.o_norm.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.b_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.b_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.b_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.v_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.v_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.k_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.k_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.q_proj.scales": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.q_proj.biases": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attention_res.proj_weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attention_res.norm_weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.mlp_res.proj_weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.mlp_res.norm_weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.e_score_correction_bias": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.q_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.k_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.self_attn.v_conv.conv.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00024-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.up_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.up_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.down_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.30.block_sparse_moe.switch_mlp.down_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.kv_a_proj_with_mqa.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.kv_a_proj_with_mqa.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.kv_a_proj_with_mqa.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_b_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_b_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_b_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_norm.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_up_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_up_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_up_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_down_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_down_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.routed_expert_down_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.up_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.up_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.up_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.down_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.down_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.down_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.gate_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.gate_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.shared_experts.gate_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.gate.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.gate.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.gate.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.g_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.g_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.g_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.o_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.o_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.kv_a_layernorm.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_a_layernorm.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_a_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_a_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.q_a_proj.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attention_res.proj_weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attention_res.norm_weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.mlp_res.proj_weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.mlp_res.norm_weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.e_score_correction_bias": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.embed_q.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.embed_q.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.embed_q.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.unembed_out.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.unembed_out.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.self_attn.unembed_out.biases": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.up_proj.weight": "model-00025-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.up_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.down_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.31.block_sparse_moe.switch_mlp.down_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_norm.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_up_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_up_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_up_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_down_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_down_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.routed_expert_down_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.up_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.up_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.up_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.down_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.down_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.down_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.gate_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.gate_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.shared_experts.gate_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.gate.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.gate.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.gate.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.g_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.g_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.g_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.dt_bias": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_b_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_b_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_b_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_a_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_a_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.f_a_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.A_log": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.o_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.o_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.o_norm.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.b_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.b_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.b_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.v_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.v_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.k_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.k_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.q_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.q_proj.biases": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attention_res.proj_weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attention_res.norm_weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.mlp_res.proj_weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.mlp_res.norm_weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.e_score_correction_bias": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.q_conv.conv.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.k_conv.conv.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.self_attn.v_conv.conv.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.up_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.up_proj.scales": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.down_proj.weight": "model-00026-of-00075.safetensors",
+    "model.layers.32.block_sparse_moe.switch_mlp.down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_norm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_up_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_up_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_up_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_down_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.routed_expert_down_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.up_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.up_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.up_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.down_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.down_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.gate_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.gate_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.shared_experts.gate_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.gate.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.gate.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.gate.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.g_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.g_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.g_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.dt_bias": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_b_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_b_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_b_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_a_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_a_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.f_a_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.A_log": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.o_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.o_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.o_norm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.b_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.b_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.b_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.v_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.v_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.k_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.k_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.q_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.q_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attention_res.proj_weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attention_res.norm_weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.mlp_res.proj_weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.mlp_res.norm_weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.e_score_correction_bias": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.q_conv.conv.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.k_conv.conv.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.self_attn.v_conv.conv.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.up_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.up_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.down_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.33.block_sparse_moe.switch_mlp.down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_norm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_up_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_up_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_up_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_down_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.routed_expert_down_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.up_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.up_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.up_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.down_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.down_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.down_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.gate_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.gate_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.shared_experts.gate_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.gate.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.gate.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.gate.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.g_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.g_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.g_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.dt_bias": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_b_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_b_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_b_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_a_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_a_proj.scales": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.f_a_proj.biases": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.A_log": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00027-of-00075.safetensors",
+    "model.layers.34.self_attn.o_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.o_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.o_norm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.b_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.b_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.b_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.v_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.v_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.k_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.k_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.q_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.q_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attention_res.proj_weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attention_res.norm_weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.mlp_res.proj_weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.mlp_res.norm_weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.e_score_correction_bias": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.q_conv.conv.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.k_conv.conv.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.self_attn.v_conv.conv.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.up_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.up_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.down_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.34.block_sparse_moe.switch_mlp.down_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.kv_a_proj_with_mqa.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.kv_a_proj_with_mqa.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.kv_a_proj_with_mqa.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_b_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_b_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_b_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_norm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_up_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_up_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_up_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_down_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_down_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.routed_expert_down_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.up_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.up_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.up_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.down_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.down_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.down_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.gate_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.gate_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.shared_experts.gate_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.gate.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.gate.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.gate.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.g_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.g_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.g_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.o_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.o_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.kv_a_layernorm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_a_layernorm.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_a_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_a_proj.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.q_a_proj.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attention_res.proj_weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attention_res.norm_weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.mlp_res.proj_weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.mlp_res.norm_weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.e_score_correction_bias": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.embed_q.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.embed_q.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.embed_q.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.unembed_out.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.unembed_out.scales": "model-00028-of-00075.safetensors",
+    "model.layers.35.self_attn.unembed_out.biases": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00028-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.up_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.up_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.down_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.35.block_sparse_moe.switch_mlp.down_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_norm.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_up_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_up_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_up_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_down_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_down_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.routed_expert_down_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.up_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.up_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.up_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.down_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.down_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.down_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.gate_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.gate_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.shared_experts.gate_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.gate.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.gate.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.gate.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.g_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.g_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.g_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.dt_bias": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_b_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_b_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_b_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_a_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_a_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.f_a_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.A_log": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.o_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.o_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.o_norm.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.b_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.b_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.b_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.v_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.v_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.k_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.k_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.q_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.q_proj.biases": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attention_res.proj_weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attention_res.norm_weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.mlp_res.proj_weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.mlp_res.norm_weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.e_score_correction_bias": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.q_conv.conv.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.k_conv.conv.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.self_attn.v_conv.conv.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.up_proj.weight": "model-00029-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.up_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.down_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.36.block_sparse_moe.switch_mlp.down_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_norm.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_up_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_up_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_up_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_down_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_down_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.routed_expert_down_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.up_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.up_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.up_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.down_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.down_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.down_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.gate_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.gate_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.shared_experts.gate_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.gate.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.gate.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.gate.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.g_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.g_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.g_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.dt_bias": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_b_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_b_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_b_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_a_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_a_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.f_a_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.A_log": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.o_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.o_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.o_norm.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.b_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.b_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.b_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.v_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.v_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.k_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.k_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.q_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.q_proj.biases": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attention_res.proj_weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attention_res.norm_weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.mlp_res.proj_weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.mlp_res.norm_weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.e_score_correction_bias": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.q_conv.conv.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.k_conv.conv.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.self_attn.v_conv.conv.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.up_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.up_proj.scales": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.down_proj.weight": "model-00030-of-00075.safetensors",
+    "model.layers.37.block_sparse_moe.switch_mlp.down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_norm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_up_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_up_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_up_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_down_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.routed_expert_down_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.up_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.up_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.up_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.down_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.down_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.gate_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.gate_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.shared_experts.gate_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.gate.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.gate.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.gate.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.g_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.g_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.g_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.dt_bias": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_b_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_b_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_b_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_a_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_a_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.f_a_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.A_log": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.o_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.o_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.o_norm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.b_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.b_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.b_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.v_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.v_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.k_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.k_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.q_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.q_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attention_res.proj_weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attention_res.norm_weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.mlp_res.proj_weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.mlp_res.norm_weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.e_score_correction_bias": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.q_conv.conv.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.k_conv.conv.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.self_attn.v_conv.conv.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.up_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.up_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.down_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.38.block_sparse_moe.switch_mlp.down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.kv_a_proj_with_mqa.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.kv_a_proj_with_mqa.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.kv_a_proj_with_mqa.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.q_b_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.q_b_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.q_b_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_norm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_up_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_up_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_up_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_down_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.routed_expert_down_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.up_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.up_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.up_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.down_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.down_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.down_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.gate_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.gate_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.shared_experts.gate_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.gate.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.gate.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.gate.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.g_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.g_proj.scales": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.g_proj.biases": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00031-of-00075.safetensors",
+    "model.layers.39.self_attn.o_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.o_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.kv_a_layernorm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.q_a_layernorm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.q_a_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.q_a_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.q_a_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attention_res.proj_weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attention_res.norm_weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.mlp_res.proj_weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.mlp_res.norm_weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.e_score_correction_bias": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.embed_q.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.embed_q.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.embed_q.biases": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.unembed_out.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.unembed_out.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.self_attn.unembed_out.biases": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.up_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.up_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.down_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.39.block_sparse_moe.switch_mlp.down_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_norm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_up_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_up_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_up_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_down_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_down_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.routed_expert_down_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.up_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.up_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.up_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.down_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.down_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.down_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.gate_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.gate_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.shared_experts.gate_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.gate.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.gate.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.gate.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.g_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.g_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.g_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.dt_bias": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_b_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_b_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_b_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_a_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_a_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.f_a_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.A_log": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.o_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.o_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.o_norm.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.b_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.b_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.b_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.v_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.v_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.k_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.k_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.q_proj.scales": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.q_proj.biases": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attention_res.proj_weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attention_res.norm_weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.mlp_res.proj_weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.mlp_res.norm_weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.e_score_correction_bias": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.q_conv.conv.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.k_conv.conv.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.self_attn.v_conv.conv.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00032-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.up_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.up_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.down_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.40.block_sparse_moe.switch_mlp.down_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_norm.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_up_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_up_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_up_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_down_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_down_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.routed_expert_down_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.up_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.up_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.up_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.down_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.down_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.down_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.gate_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.gate_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.shared_experts.gate_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.gate.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.gate.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.gate.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.g_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.g_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.g_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.dt_bias": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_b_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_b_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_b_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_a_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_a_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.f_a_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.A_log": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.o_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.o_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.o_norm.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.b_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.b_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.b_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.v_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.v_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.k_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.k_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.q_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.q_proj.biases": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attention_res.proj_weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attention_res.norm_weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.mlp_res.proj_weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.mlp_res.norm_weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.e_score_correction_bias": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.q_conv.conv.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.k_conv.conv.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.self_attn.v_conv.conv.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.up_proj.weight": "model-00033-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.up_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.down_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.41.block_sparse_moe.switch_mlp.down_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_norm.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_up_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_up_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_up_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_down_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_down_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.routed_expert_down_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.up_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.up_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.up_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.down_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.down_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.down_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.gate_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.gate_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.shared_experts.gate_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.gate.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.gate.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.gate.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.g_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.g_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.g_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.dt_bias": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_b_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_b_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_b_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_a_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_a_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.f_a_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.A_log": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.o_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.o_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.o_norm.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.b_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.b_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.b_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.v_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.v_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.k_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.k_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.q_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.q_proj.biases": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attention_res.proj_weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attention_res.norm_weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.mlp_res.proj_weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.mlp_res.norm_weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.e_score_correction_bias": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.q_conv.conv.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.k_conv.conv.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.self_attn.v_conv.conv.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.up_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.up_proj.scales": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.down_proj.weight": "model-00034-of-00075.safetensors",
+    "model.layers.42.block_sparse_moe.switch_mlp.down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.kv_a_proj_with_mqa.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.kv_a_proj_with_mqa.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.kv_a_proj_with_mqa.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_b_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_b_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_b_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_norm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_up_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_up_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_up_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_down_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.routed_expert_down_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.up_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.up_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.up_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.down_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.down_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.gate_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.gate_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.shared_experts.gate_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.gate.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.gate.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.gate.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.g_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.g_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.g_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.o_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.o_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.kv_a_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_a_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_a_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_a_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.q_a_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attention_res.proj_weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attention_res.norm_weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.mlp_res.proj_weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.mlp_res.norm_weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.e_score_correction_bias": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.embed_q.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.embed_q.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.embed_q.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.unembed_out.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.unembed_out.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.self_attn.unembed_out.biases": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.up_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.up_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.down_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.43.block_sparse_moe.switch_mlp.down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_norm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_up_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_up_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_up_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_down_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.routed_expert_down_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.up_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.up_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.up_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.down_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.down_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.down_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.gate_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.gate_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.shared_experts.gate_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.gate.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.gate.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.gate.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.g_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.g_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.g_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.dt_bias": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_b_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_b_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_b_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_a_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_a_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.f_a_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.A_log": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.o_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.o_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.o_norm.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.b_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.b_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.b_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.v_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.v_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.k_proj.scales": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.k_proj.biases": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00035-of-00075.safetensors",
+    "model.layers.44.self_attn.q_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attn.q_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attention_res.proj_weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attention_res.norm_weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.mlp_res.proj_weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.mlp_res.norm_weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.e_score_correction_bias": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attn.q_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attn.k_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.self_attn.v_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.up_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.up_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.down_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.44.block_sparse_moe.switch_mlp.down_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_norm.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_up_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_up_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_up_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_down_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_down_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.routed_expert_down_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.up_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.up_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.up_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.down_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.down_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.down_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.gate_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.gate_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.shared_experts.gate_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.gate.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.gate.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.gate.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.g_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.g_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.g_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.dt_bias": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_b_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_b_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_b_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_a_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_a_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.f_a_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.A_log": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.o_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.o_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.o_norm.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.b_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.b_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.b_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.v_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.v_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.k_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.k_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.q_proj.scales": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.q_proj.biases": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attention_res.proj_weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attention_res.norm_weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.mlp_res.proj_weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.mlp_res.norm_weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.e_score_correction_bias": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.q_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.k_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.self_attn.v_conv.conv.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00036-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.up_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.up_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.down_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.45.block_sparse_moe.switch_mlp.down_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_norm.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_up_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_up_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_up_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_down_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_down_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.routed_expert_down_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.up_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.up_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.up_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.down_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.down_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.down_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.gate_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.gate_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.shared_experts.gate_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.gate.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.gate.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.gate.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.g_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.g_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.g_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.dt_bias": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_b_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_b_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_b_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_a_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_a_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.f_a_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.A_log": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.o_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.o_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.o_norm.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.b_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.b_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.b_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.v_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.v_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.k_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.k_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.q_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.q_proj.biases": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attention_res.proj_weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attention_res.norm_weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.mlp_res.proj_weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.mlp_res.norm_weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.e_score_correction_bias": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.q_conv.conv.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.k_conv.conv.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.self_attn.v_conv.conv.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.up_proj.weight": "model-00037-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.up_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.down_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.46.block_sparse_moe.switch_mlp.down_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.kv_a_proj_with_mqa.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.kv_a_proj_with_mqa.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.kv_a_proj_with_mqa.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_b_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_b_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_b_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_norm.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_up_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_up_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_up_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_down_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_down_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.routed_expert_down_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.up_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.up_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.up_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.down_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.down_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.down_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.gate_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.gate_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.shared_experts.gate_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.gate.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.gate.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.gate.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.g_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.g_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.g_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.o_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.o_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.kv_a_layernorm.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_a_layernorm.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_a_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_a_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.q_a_proj.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attention_res.proj_weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attention_res.norm_weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.mlp_res.proj_weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.mlp_res.norm_weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.e_score_correction_bias": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.embed_q.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.embed_q.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.embed_q.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.unembed_out.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.unembed_out.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.self_attn.unembed_out.biases": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.up_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.up_proj.scales": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.down_proj.weight": "model-00038-of-00075.safetensors",
+    "model.layers.47.block_sparse_moe.switch_mlp.down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_norm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_up_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_up_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_up_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_down_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.routed_expert_down_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.up_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.up_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.up_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.down_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.down_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.gate_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.gate_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.shared_experts.gate_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.gate.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.gate.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.gate.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.g_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.g_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.g_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.dt_bias": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_b_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_b_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_b_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_a_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_a_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.f_a_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.A_log": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.o_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.o_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.o_norm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.b_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.b_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.b_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.v_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.v_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.k_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.k_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.q_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.q_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attention_res.proj_weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attention_res.norm_weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.mlp_res.proj_weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.mlp_res.norm_weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.e_score_correction_bias": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.q_conv.conv.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.k_conv.conv.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.self_attn.v_conv.conv.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.up_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.up_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.down_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.48.block_sparse_moe.switch_mlp.down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_norm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_up_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_up_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_up_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_down_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.routed_expert_down_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.up_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.up_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.up_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.down_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.down_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.down_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.gate_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.gate_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.shared_experts.gate_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.gate.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.gate.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.gate.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.g_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.g_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.g_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.dt_bias": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_b_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_b_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_b_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_a_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_a_proj.scales": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.f_a_proj.biases": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.A_log": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00039-of-00075.safetensors",
+    "model.layers.49.self_attn.o_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.o_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.o_norm.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.b_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.b_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.b_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.v_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.v_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.k_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.k_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.q_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.q_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attention_res.proj_weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attention_res.norm_weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.mlp_res.proj_weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.mlp_res.norm_weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.e_score_correction_bias": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.q_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.k_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.self_attn.v_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.up_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.up_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.down_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.49.block_sparse_moe.switch_mlp.down_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_norm.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_up_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_up_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_up_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_down_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_down_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.routed_expert_down_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.up_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.up_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.up_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.down_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.down_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.down_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.gate_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.gate_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.shared_experts.gate_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.gate.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.gate.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.gate.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.g_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.g_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.g_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.dt_bias": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_b_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_b_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_b_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_a_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_a_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.f_a_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.A_log": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.o_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.o_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.o_norm.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.b_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.b_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.b_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.v_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.v_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.k_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.k_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.q_proj.scales": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.q_proj.biases": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attention_res.proj_weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attention_res.norm_weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.mlp_res.proj_weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.mlp_res.norm_weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.e_score_correction_bias": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.q_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.k_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.self_attn.v_conv.conv.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00040-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.up_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.up_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.down_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.50.block_sparse_moe.switch_mlp.down_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.kv_a_proj_with_mqa.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.kv_a_proj_with_mqa.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.kv_a_proj_with_mqa.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_b_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_b_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_b_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_norm.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_up_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_up_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_up_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_down_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_down_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.routed_expert_down_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.up_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.up_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.up_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.down_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.down_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.down_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.gate_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.gate_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.shared_experts.gate_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.gate.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.gate.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.gate.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.g_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.g_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.g_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.o_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.o_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.kv_a_layernorm.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_a_layernorm.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_a_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_a_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.q_a_proj.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attention_res.proj_weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attention_res.norm_weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.mlp_res.proj_weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.mlp_res.norm_weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.e_score_correction_bias": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.embed_q.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.embed_q.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.embed_q.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.unembed_out.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.unembed_out.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.self_attn.unembed_out.biases": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.up_proj.weight": "model-00041-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.up_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.down_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.51.block_sparse_moe.switch_mlp.down_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_norm.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_up_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_up_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_up_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_down_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_down_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.routed_expert_down_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.up_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.up_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.up_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.down_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.down_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.down_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.gate_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.gate_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.shared_experts.gate_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.gate.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.gate.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.gate.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.g_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.g_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.g_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.dt_bias": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_b_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_b_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_b_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_a_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_a_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.f_a_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.A_log": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.o_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.o_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.o_norm.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.b_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.b_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.b_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.v_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.v_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.k_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.k_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.q_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.q_proj.biases": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attention_res.proj_weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attention_res.norm_weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.mlp_res.proj_weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.mlp_res.norm_weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.e_score_correction_bias": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.q_conv.conv.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.k_conv.conv.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.self_attn.v_conv.conv.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.up_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.up_proj.scales": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.down_proj.weight": "model-00042-of-00075.safetensors",
+    "model.layers.52.block_sparse_moe.switch_mlp.down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_norm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_up_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_up_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_up_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_down_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.routed_expert_down_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.up_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.up_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.up_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.down_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.down_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.gate_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.gate_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.shared_experts.gate_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.gate.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.gate.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.gate.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.g_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.g_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.g_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.dt_bias": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_b_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_b_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_b_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_a_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_a_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.f_a_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.A_log": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.o_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.o_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.o_norm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.b_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.b_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.b_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.v_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.v_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.k_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.k_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.q_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.q_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attention_res.proj_weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attention_res.norm_weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.mlp_res.proj_weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.mlp_res.norm_weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.e_score_correction_bias": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.q_conv.conv.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.k_conv.conv.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.self_attn.v_conv.conv.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.up_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.up_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.down_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.53.block_sparse_moe.switch_mlp.down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_norm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_up_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_up_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_up_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_down_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.routed_expert_down_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.up_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.up_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.up_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.down_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.down_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.down_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.gate_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.gate_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.shared_experts.gate_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.gate.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.gate.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.gate.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.g_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.g_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.g_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.dt_bias": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_b_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_b_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_b_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_a_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_a_proj.scales": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.f_a_proj.biases": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.A_log": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00043-of-00075.safetensors",
+    "model.layers.54.self_attn.o_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.o_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.o_norm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.b_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.b_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.b_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.v_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.v_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.k_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.k_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.q_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.q_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attention_res.proj_weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attention_res.norm_weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.mlp_res.proj_weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.mlp_res.norm_weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.e_score_correction_bias": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.q_conv.conv.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.k_conv.conv.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.self_attn.v_conv.conv.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.up_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.up_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.down_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.54.block_sparse_moe.switch_mlp.down_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.kv_a_proj_with_mqa.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.kv_a_proj_with_mqa.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.kv_a_proj_with_mqa.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_b_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_b_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_b_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_norm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_up_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_up_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_up_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_down_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_down_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.routed_expert_down_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.up_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.up_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.up_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.down_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.down_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.down_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.gate_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.gate_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.shared_experts.gate_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.gate.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.gate.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.gate.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.g_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.g_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.g_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.o_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.o_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.kv_a_layernorm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_a_layernorm.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_a_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_a_proj.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.q_a_proj.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attention_res.proj_weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attention_res.norm_weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.mlp_res.proj_weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.mlp_res.norm_weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.e_score_correction_bias": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.embed_q.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.embed_q.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.embed_q.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.unembed_out.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.unembed_out.scales": "model-00044-of-00075.safetensors",
+    "model.layers.55.self_attn.unembed_out.biases": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00044-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.up_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.up_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.down_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.55.block_sparse_moe.switch_mlp.down_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_norm.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_up_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_up_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_up_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_down_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_down_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.routed_expert_down_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.up_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.up_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.up_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.down_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.down_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.down_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.gate_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.gate_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.shared_experts.gate_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.gate.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.gate.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.gate.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.g_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.g_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.g_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.dt_bias": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_b_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_b_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_b_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_a_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_a_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.f_a_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.A_log": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.o_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.o_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.o_norm.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.b_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.b_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.b_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.v_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.v_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.k_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.k_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.q_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.q_proj.biases": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attention_res.proj_weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attention_res.norm_weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.mlp_res.proj_weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.mlp_res.norm_weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.e_score_correction_bias": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.q_conv.conv.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.k_conv.conv.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.self_attn.v_conv.conv.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.up_proj.weight": "model-00045-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.up_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.down_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.56.block_sparse_moe.switch_mlp.down_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_norm.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_up_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_up_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_up_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_down_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_down_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.routed_expert_down_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.up_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.up_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.up_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.down_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.down_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.down_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.gate_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.gate_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.shared_experts.gate_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.gate.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.gate.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.gate.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.g_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.g_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.g_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.dt_bias": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_b_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_b_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_b_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_a_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_a_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.f_a_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.A_log": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.o_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.o_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.o_norm.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.b_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.b_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.b_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.v_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.v_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.k_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.k_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.q_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.q_proj.biases": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attention_res.proj_weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attention_res.norm_weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.mlp_res.proj_weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.mlp_res.norm_weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.e_score_correction_bias": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.q_conv.conv.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.k_conv.conv.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.self_attn.v_conv.conv.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.up_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.up_proj.scales": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.down_proj.weight": "model-00046-of-00075.safetensors",
+    "model.layers.57.block_sparse_moe.switch_mlp.down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_norm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_up_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_up_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_up_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_down_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.routed_expert_down_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.up_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.up_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.up_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.down_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.down_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.gate_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.gate_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.shared_experts.gate_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.gate.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.gate.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.gate.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.g_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.g_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.g_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.dt_bias": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_b_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_b_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_b_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_a_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_a_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.f_a_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.A_log": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.o_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.o_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.o_norm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.b_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.b_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.b_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.v_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.v_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.k_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.k_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.q_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.q_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attention_res.proj_weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attention_res.norm_weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.mlp_res.proj_weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.mlp_res.norm_weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.e_score_correction_bias": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.q_conv.conv.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.k_conv.conv.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.self_attn.v_conv.conv.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.up_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.up_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.down_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.58.block_sparse_moe.switch_mlp.down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.kv_a_proj_with_mqa.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.kv_a_proj_with_mqa.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.kv_a_proj_with_mqa.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.q_b_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.q_b_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.q_b_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_norm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_up_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_up_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_up_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_down_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.routed_expert_down_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.up_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.up_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.up_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.down_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.down_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.down_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.gate_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.gate_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.shared_experts.gate_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.gate.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.gate.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.gate.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.g_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.g_proj.scales": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.g_proj.biases": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00047-of-00075.safetensors",
+    "model.layers.59.self_attn.o_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.o_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.kv_a_layernorm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.q_a_layernorm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.q_a_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.q_a_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.q_a_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attention_res.proj_weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attention_res.norm_weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.mlp_res.proj_weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.mlp_res.norm_weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.e_score_correction_bias": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.embed_q.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.embed_q.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.embed_q.biases": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.unembed_out.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.unembed_out.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.self_attn.unembed_out.biases": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.up_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.up_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.down_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.59.block_sparse_moe.switch_mlp.down_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_norm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_up_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_up_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_up_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_down_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_down_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.routed_expert_down_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.up_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.up_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.up_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.down_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.down_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.down_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.gate_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.gate_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.shared_experts.gate_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.gate.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.gate.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.gate.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.post_attention_layernorm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.input_layernorm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.g_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.g_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.g_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.dt_bias": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_b_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_b_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_b_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_a_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_a_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.f_a_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.A_log": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.o_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.o_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.o_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.o_norm.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.b_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.b_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.b_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.v_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.v_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.v_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.k_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.k_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.k_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.q_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.q_proj.scales": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.q_proj.biases": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attention_res.proj_weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attention_res.norm_weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.mlp_res.proj_weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.mlp_res.norm_weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.e_score_correction_bias": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.q_conv.conv.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.k_conv.conv.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.self_attn.v_conv.conv.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00048-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.up_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.up_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.down_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.60.block_sparse_moe.switch_mlp.down_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_norm.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_up_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_up_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_up_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_down_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_down_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.routed_expert_down_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.up_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.up_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.up_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.down_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.down_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.down_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.gate_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.gate_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.shared_experts.gate_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.gate.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.gate.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.gate.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.post_attention_layernorm.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.input_layernorm.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.g_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.g_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.g_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.dt_bias": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_b_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_b_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_b_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_a_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_a_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.f_a_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.A_log": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.o_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.o_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.o_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.o_norm.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.b_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.b_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.b_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.v_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.v_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.v_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.k_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.k_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.k_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.q_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.q_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.q_proj.biases": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attention_res.proj_weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attention_res.norm_weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.mlp_res.proj_weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.mlp_res.norm_weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.e_score_correction_bias": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.q_conv.conv.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.k_conv.conv.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.self_attn.v_conv.conv.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.up_proj.weight": "model-00049-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.up_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.down_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.61.block_sparse_moe.switch_mlp.down_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_norm.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_up_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_up_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_up_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_down_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_down_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.routed_expert_down_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.up_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.up_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.up_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.down_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.down_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.down_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.gate_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.gate_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.shared_experts.gate_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.gate.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.gate.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.gate.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.post_attention_layernorm.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.input_layernorm.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.g_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.g_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.g_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.dt_bias": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_b_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_b_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_b_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_a_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_a_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.f_a_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.A_log": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.o_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.o_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.o_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.o_norm.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.b_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.b_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.b_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.v_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.v_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.v_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.k_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.k_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.k_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.q_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.q_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.q_proj.biases": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attention_res.proj_weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attention_res.norm_weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.mlp_res.proj_weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.mlp_res.norm_weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.e_score_correction_bias": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.q_conv.conv.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.k_conv.conv.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.self_attn.v_conv.conv.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.up_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.up_proj.scales": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.down_proj.weight": "model-00050-of-00075.safetensors",
+    "model.layers.62.block_sparse_moe.switch_mlp.down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.kv_a_proj_with_mqa.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.kv_a_proj_with_mqa.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.kv_a_proj_with_mqa.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_b_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_b_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_b_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_norm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_up_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_up_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_up_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_down_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.routed_expert_down_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.up_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.up_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.up_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.down_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.down_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.gate_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.gate_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.shared_experts.gate_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.gate.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.gate.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.gate.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.post_attention_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.input_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.g_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.g_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.g_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.o_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.o_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.o_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.kv_a_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_a_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_a_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_a_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.q_a_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attention_res.proj_weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attention_res.norm_weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.mlp_res.proj_weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.mlp_res.norm_weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.e_score_correction_bias": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.embed_q.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.embed_q.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.embed_q.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.unembed_out.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.unembed_out.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.self_attn.unembed_out.biases": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.up_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.up_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.down_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.63.block_sparse_moe.switch_mlp.down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_norm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_up_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_up_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_up_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_down_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.routed_expert_down_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.up_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.up_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.up_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.down_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.down_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.down_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.gate_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.gate_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.shared_experts.gate_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.gate.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.gate.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.gate.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.post_attention_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.input_layernorm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.g_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.g_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.g_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.dt_bias": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_b_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_b_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_b_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_a_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_a_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.f_a_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.A_log": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.o_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.o_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.o_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.o_norm.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.b_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.b_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.b_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.v_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.v_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.v_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.k_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.k_proj.scales": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.k_proj.biases": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.q_proj.weight": "model-00051-of-00075.safetensors",
+    "model.layers.64.self_attn.q_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attn.q_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attention_res.proj_weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attention_res.norm_weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.mlp_res.proj_weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.mlp_res.norm_weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.e_score_correction_bias": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attn.q_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attn.k_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.self_attn.v_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.up_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.up_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.down_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.64.block_sparse_moe.switch_mlp.down_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_norm.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_up_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_up_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_up_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_down_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_down_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.routed_expert_down_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.up_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.up_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.up_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.down_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.down_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.down_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.gate_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.gate_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.shared_experts.gate_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.gate.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.gate.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.gate.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.post_attention_layernorm.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.input_layernorm.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.g_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.g_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.g_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.dt_bias": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_b_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_b_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_b_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_a_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_a_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.f_a_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.A_log": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.o_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.o_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.o_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.o_norm.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.b_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.b_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.b_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.v_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.v_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.v_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.k_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.k_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.k_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.q_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.q_proj.scales": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.q_proj.biases": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attention_res.proj_weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attention_res.norm_weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.mlp_res.proj_weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.mlp_res.norm_weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.e_score_correction_bias": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.q_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.k_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.self_attn.v_conv.conv.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00052-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.up_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.up_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.down_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.65.block_sparse_moe.switch_mlp.down_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_norm.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_up_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_up_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_up_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_down_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_down_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.routed_expert_down_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.up_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.up_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.up_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.down_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.down_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.down_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.gate_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.gate_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.shared_experts.gate_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.gate.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.gate.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.gate.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.post_attention_layernorm.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.input_layernorm.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.g_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.g_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.g_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.dt_bias": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_b_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_b_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_b_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_a_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_a_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.f_a_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.A_log": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.o_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.o_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.o_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.o_norm.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.b_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.b_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.b_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.v_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.v_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.v_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.k_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.k_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.k_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.q_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.q_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.q_proj.biases": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attention_res.proj_weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attention_res.norm_weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.mlp_res.proj_weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.mlp_res.norm_weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.e_score_correction_bias": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.q_conv.conv.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.k_conv.conv.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.self_attn.v_conv.conv.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.up_proj.weight": "model-00053-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.up_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.down_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.66.block_sparse_moe.switch_mlp.down_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.kv_a_proj_with_mqa.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.kv_a_proj_with_mqa.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.kv_a_proj_with_mqa.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_b_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_b_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_b_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_norm.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_up_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_up_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_up_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_down_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_down_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.routed_expert_down_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.up_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.up_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.up_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.down_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.down_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.down_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.gate_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.gate_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.shared_experts.gate_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.gate.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.gate.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.gate.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.post_attention_layernorm.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.input_layernorm.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.g_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.g_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.g_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.o_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.o_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.o_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.kv_a_layernorm.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_a_layernorm.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_a_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_a_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.q_a_proj.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attention_res.proj_weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attention_res.norm_weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.mlp_res.proj_weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.mlp_res.norm_weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.e_score_correction_bias": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.embed_q.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.embed_q.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.embed_q.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.unembed_out.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.unembed_out.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.self_attn.unembed_out.biases": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.up_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.up_proj.scales": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.down_proj.weight": "model-00054-of-00075.safetensors",
+    "model.layers.67.block_sparse_moe.switch_mlp.down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_norm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_up_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_up_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_up_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_down_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.routed_expert_down_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.up_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.up_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.up_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.down_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.down_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.gate_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.gate_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.shared_experts.gate_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.gate.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.gate.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.gate.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.post_attention_layernorm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.input_layernorm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.g_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.g_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.g_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.dt_bias": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_b_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_b_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_b_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_a_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_a_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.f_a_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.A_log": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.o_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.o_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.o_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.o_norm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.b_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.b_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.b_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.v_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.v_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.v_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.k_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.k_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.k_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.q_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.q_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.q_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attention_res.proj_weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attention_res.norm_weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.mlp_res.proj_weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.mlp_res.norm_weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.e_score_correction_bias": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.q_conv.conv.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.k_conv.conv.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.self_attn.v_conv.conv.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.up_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.up_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.down_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.68.block_sparse_moe.switch_mlp.down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_norm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_up_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_up_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_up_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_down_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.routed_expert_down_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.up_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.up_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.up_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.down_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.down_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.down_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.gate_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.gate_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.shared_experts.gate_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.gate.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.gate.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.gate.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.post_attention_layernorm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.input_layernorm.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.g_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.g_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.g_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.dt_bias": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_b_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_b_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_b_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_a_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_a_proj.scales": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.f_a_proj.biases": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.A_log": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.o_proj.weight": "model-00055-of-00075.safetensors",
+    "model.layers.69.self_attn.o_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.o_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.o_norm.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.b_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.b_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.b_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.v_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.v_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.v_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.k_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.k_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.k_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.q_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.q_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.q_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attention_res.proj_weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attention_res.norm_weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.mlp_res.proj_weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.mlp_res.norm_weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.e_score_correction_bias": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.q_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.k_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.self_attn.v_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.up_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.up_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.down_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.69.block_sparse_moe.switch_mlp.down_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_norm.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_up_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_up_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_up_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_down_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_down_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.routed_expert_down_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.up_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.up_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.up_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.down_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.down_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.down_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.gate_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.gate_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.shared_experts.gate_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.gate.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.gate.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.gate.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.post_attention_layernorm.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.input_layernorm.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.g_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.g_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.g_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.dt_bias": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_b_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_b_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_b_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_a_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_a_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.f_a_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.A_log": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.o_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.o_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.o_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.o_norm.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.b_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.b_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.b_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.v_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.v_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.v_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.k_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.k_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.k_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.q_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.q_proj.scales": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.q_proj.biases": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attention_res.proj_weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attention_res.norm_weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.mlp_res.proj_weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.mlp_res.norm_weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.e_score_correction_bias": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.q_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.k_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.self_attn.v_conv.conv.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00056-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.up_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.up_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.down_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.70.block_sparse_moe.switch_mlp.down_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.kv_a_proj_with_mqa.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.kv_a_proj_with_mqa.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.kv_a_proj_with_mqa.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_b_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_b_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_b_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_norm.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_up_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_up_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_up_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_down_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_down_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.routed_expert_down_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.up_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.up_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.up_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.down_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.down_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.down_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.gate_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.gate_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.shared_experts.gate_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.gate.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.gate.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.gate.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.post_attention_layernorm.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.input_layernorm.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.g_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.g_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.g_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.o_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.o_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.o_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.kv_a_layernorm.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_a_layernorm.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_a_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_a_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.q_a_proj.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attention_res.proj_weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attention_res.norm_weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.mlp_res.proj_weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.mlp_res.norm_weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.e_score_correction_bias": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.embed_q.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.embed_q.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.embed_q.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.unembed_out.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.unembed_out.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.self_attn.unembed_out.biases": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.up_proj.weight": "model-00057-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.up_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.down_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.71.block_sparse_moe.switch_mlp.down_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_norm.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_up_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_up_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_up_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_down_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_down_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.routed_expert_down_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.up_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.up_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.up_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.down_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.down_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.down_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.gate_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.gate_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.shared_experts.gate_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.gate.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.gate.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.gate.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.post_attention_layernorm.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.input_layernorm.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.g_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.g_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.g_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.dt_bias": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_b_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_b_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_b_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_a_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_a_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.f_a_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.A_log": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.o_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.o_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.o_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.o_norm.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.b_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.b_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.b_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.v_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.v_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.v_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.k_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.k_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.k_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.q_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.q_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.q_proj.biases": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attention_res.proj_weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attention_res.norm_weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.mlp_res.proj_weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.mlp_res.norm_weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.e_score_correction_bias": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.q_conv.conv.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.k_conv.conv.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.self_attn.v_conv.conv.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.up_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.up_proj.scales": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.down_proj.weight": "model-00058-of-00075.safetensors",
+    "model.layers.72.block_sparse_moe.switch_mlp.down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_norm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_up_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_up_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_up_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_down_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.routed_expert_down_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.up_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.up_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.up_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.down_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.down_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.gate_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.gate_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.shared_experts.gate_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.gate.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.gate.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.gate.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.post_attention_layernorm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.input_layernorm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.g_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.g_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.g_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.dt_bias": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_b_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_b_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_b_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_a_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_a_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.f_a_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.A_log": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.o_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.o_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.o_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.o_norm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.b_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.b_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.b_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.v_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.v_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.v_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.k_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.k_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.k_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.q_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.q_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.q_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attention_res.proj_weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attention_res.norm_weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.mlp_res.proj_weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.mlp_res.norm_weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.e_score_correction_bias": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.q_conv.conv.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.k_conv.conv.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.self_attn.v_conv.conv.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.up_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.up_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.down_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.73.block_sparse_moe.switch_mlp.down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_norm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_up_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_up_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_up_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_down_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.routed_expert_down_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.up_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.up_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.up_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.down_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.down_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.down_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.gate_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.gate_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.shared_experts.gate_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.gate.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.gate.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.gate.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.post_attention_layernorm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.input_layernorm.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.g_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.g_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.g_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.dt_bias": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_b_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_b_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_b_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_a_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_a_proj.scales": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.f_a_proj.biases": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.A_log": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.o_proj.weight": "model-00059-of-00075.safetensors",
+    "model.layers.74.self_attn.o_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.o_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.o_norm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.b_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.b_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.b_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.v_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.v_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.v_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.k_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.k_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.k_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.q_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.q_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.q_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attention_res.proj_weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attention_res.norm_weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.mlp_res.proj_weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.mlp_res.norm_weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.e_score_correction_bias": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.q_conv.conv.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.k_conv.conv.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.self_attn.v_conv.conv.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.up_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.up_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.down_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.74.block_sparse_moe.switch_mlp.down_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.kv_a_proj_with_mqa.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.kv_a_proj_with_mqa.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.kv_a_proj_with_mqa.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_b_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_b_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_b_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_norm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_up_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_up_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_up_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_down_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_down_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.routed_expert_down_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.up_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.up_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.up_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.down_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.down_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.down_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.gate_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.gate_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.shared_experts.gate_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.gate.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.gate.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.gate.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.post_attention_layernorm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.input_layernorm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.g_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.g_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.g_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.o_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.o_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.o_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.kv_a_layernorm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_a_layernorm.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_a_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_a_proj.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.q_a_proj.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attention_res.proj_weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attention_res.norm_weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.mlp_res.proj_weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.mlp_res.norm_weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.e_score_correction_bias": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.embed_q.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.embed_q.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.embed_q.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.unembed_out.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.unembed_out.scales": "model-00060-of-00075.safetensors",
+    "model.layers.75.self_attn.unembed_out.biases": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00060-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.up_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.up_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.down_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.75.block_sparse_moe.switch_mlp.down_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_norm.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_up_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_up_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_up_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_down_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_down_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.routed_expert_down_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.up_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.up_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.up_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.down_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.down_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.down_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.gate_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.gate_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.shared_experts.gate_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.gate.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.gate.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.gate.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.post_attention_layernorm.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.input_layernorm.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.g_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.g_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.g_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.dt_bias": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_b_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_b_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_b_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_a_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_a_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.f_a_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.A_log": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.o_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.o_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.o_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.o_norm.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.b_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.b_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.b_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.v_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.v_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.v_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.k_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.k_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.k_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.q_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.q_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.q_proj.biases": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attention_res.proj_weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attention_res.norm_weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.mlp_res.proj_weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.mlp_res.norm_weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.e_score_correction_bias": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.q_conv.conv.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.k_conv.conv.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.self_attn.v_conv.conv.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.up_proj.weight": "model-00061-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.up_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.down_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.76.block_sparse_moe.switch_mlp.down_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_norm.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_up_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_up_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_up_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_down_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_down_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.routed_expert_down_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.up_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.up_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.up_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.down_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.down_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.down_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.gate_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.gate_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.shared_experts.gate_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.gate.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.gate.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.gate.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.post_attention_layernorm.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.input_layernorm.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.g_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.g_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.g_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.dt_bias": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_b_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_b_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_b_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_a_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_a_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.f_a_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.A_log": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.o_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.o_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.o_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.o_norm.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.b_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.b_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.b_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.v_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.v_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.v_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.k_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.k_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.k_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.q_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.q_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.q_proj.biases": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attention_res.proj_weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attention_res.norm_weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.mlp_res.proj_weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.mlp_res.norm_weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.e_score_correction_bias": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.q_conv.conv.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.k_conv.conv.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.self_attn.v_conv.conv.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.up_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.up_proj.scales": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.down_proj.weight": "model-00062-of-00075.safetensors",
+    "model.layers.77.block_sparse_moe.switch_mlp.down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_norm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_up_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_up_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_up_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_down_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.routed_expert_down_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.up_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.up_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.up_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.down_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.down_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.gate_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.gate_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.shared_experts.gate_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.gate.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.gate.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.gate.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.post_attention_layernorm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.input_layernorm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.g_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.g_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.g_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.dt_bias": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_b_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_b_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_b_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_a_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_a_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.f_a_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.A_log": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.o_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.o_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.o_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.o_norm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.b_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.b_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.b_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.v_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.v_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.v_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.k_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.k_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.k_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.q_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.q_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.q_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attention_res.proj_weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attention_res.norm_weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.mlp_res.proj_weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.mlp_res.norm_weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.e_score_correction_bias": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.q_conv.conv.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.k_conv.conv.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.self_attn.v_conv.conv.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.up_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.up_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.down_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.78.block_sparse_moe.switch_mlp.down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.kv_a_proj_with_mqa.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.kv_a_proj_with_mqa.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.kv_a_proj_with_mqa.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.q_b_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.q_b_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.q_b_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_norm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_up_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_up_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_up_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_down_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.routed_expert_down_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.up_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.up_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.up_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.down_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.down_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.down_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.gate_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.gate_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.shared_experts.gate_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.gate.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.gate.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.gate.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.post_attention_layernorm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.input_layernorm.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.g_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.g_proj.scales": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.g_proj.biases": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.o_proj.weight": "model-00063-of-00075.safetensors",
+    "model.layers.79.self_attn.o_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.o_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.kv_a_layernorm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.q_a_layernorm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.q_a_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.q_a_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.q_a_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attention_res.proj_weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attention_res.norm_weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.mlp_res.proj_weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.mlp_res.norm_weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.e_score_correction_bias": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.embed_q.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.embed_q.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.embed_q.biases": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.unembed_out.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.unembed_out.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.self_attn.unembed_out.biases": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.up_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.up_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.down_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.79.block_sparse_moe.switch_mlp.down_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_norm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_up_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_up_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_up_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_down_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_down_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.routed_expert_down_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.up_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.up_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.up_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.down_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.down_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.down_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.gate_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.gate_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.shared_experts.gate_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.gate.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.gate.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.gate.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.post_attention_layernorm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.input_layernorm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.g_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.g_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.g_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.dt_bias": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_b_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_b_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_b_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_a_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_a_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.f_a_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.A_log": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.o_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.o_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.o_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.o_norm.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.b_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.b_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.b_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.v_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.v_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.v_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.k_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.k_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.k_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.q_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.q_proj.scales": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.q_proj.biases": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attention_res.proj_weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attention_res.norm_weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.mlp_res.proj_weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.mlp_res.norm_weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.e_score_correction_bias": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.q_conv.conv.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.k_conv.conv.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.self_attn.v_conv.conv.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00064-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.up_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.up_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.down_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.80.block_sparse_moe.switch_mlp.down_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_norm.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_up_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_up_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_up_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_down_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_down_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.routed_expert_down_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.up_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.up_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.up_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.down_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.down_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.down_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.gate_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.gate_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.shared_experts.gate_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.gate.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.gate.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.gate.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.post_attention_layernorm.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.input_layernorm.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.g_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.g_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.g_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.dt_bias": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_b_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_b_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_b_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_a_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_a_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.f_a_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.A_log": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.o_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.o_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.o_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.o_norm.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.b_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.b_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.b_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.v_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.v_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.v_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.k_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.k_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.k_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.q_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.q_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.q_proj.biases": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attention_res.proj_weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attention_res.norm_weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.mlp_res.proj_weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.mlp_res.norm_weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.e_score_correction_bias": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.q_conv.conv.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.k_conv.conv.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.self_attn.v_conv.conv.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.up_proj.weight": "model-00065-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.up_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.down_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.81.block_sparse_moe.switch_mlp.down_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_norm.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_up_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_up_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_up_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_down_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_down_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.routed_expert_down_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.up_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.up_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.up_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.down_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.down_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.down_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.gate_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.gate_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.shared_experts.gate_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.gate.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.gate.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.gate.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.post_attention_layernorm.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.input_layernorm.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.g_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.g_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.g_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.dt_bias": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_b_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_b_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_b_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_a_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_a_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.f_a_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.A_log": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.o_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.o_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.o_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.o_norm.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.b_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.b_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.b_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.v_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.v_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.v_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.k_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.k_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.k_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.q_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.q_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.q_proj.biases": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attention_res.proj_weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attention_res.norm_weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.mlp_res.proj_weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.mlp_res.norm_weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.e_score_correction_bias": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.q_conv.conv.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.k_conv.conv.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.self_attn.v_conv.conv.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.up_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.up_proj.scales": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.down_proj.weight": "model-00066-of-00075.safetensors",
+    "model.layers.82.block_sparse_moe.switch_mlp.down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.kv_a_proj_with_mqa.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.kv_a_proj_with_mqa.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.kv_a_proj_with_mqa.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_b_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_b_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_b_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_norm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_up_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_up_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_up_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_down_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.routed_expert_down_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.up_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.up_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.up_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.down_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.down_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.gate_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.gate_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.shared_experts.gate_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.gate.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.gate.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.gate.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.post_attention_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.input_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.g_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.g_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.g_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.o_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.o_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.o_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.kv_a_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_a_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_a_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_a_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.q_a_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attention_res.proj_weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attention_res.norm_weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.mlp_res.proj_weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.mlp_res.norm_weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.e_score_correction_bias": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.embed_q.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.embed_q.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.embed_q.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.unembed_out.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.unembed_out.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.self_attn.unembed_out.biases": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.up_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.up_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.down_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.83.block_sparse_moe.switch_mlp.down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_norm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_up_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_up_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_up_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_down_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.routed_expert_down_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.up_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.up_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.up_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.down_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.down_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.down_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.gate_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.gate_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.shared_experts.gate_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.gate.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.gate.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.gate.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.post_attention_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.input_layernorm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.g_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.g_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.g_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.dt_bias": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_b_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_b_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_b_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_a_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_a_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.f_a_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.A_log": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.o_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.o_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.o_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.o_norm.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.b_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.b_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.b_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.v_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.v_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.v_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.k_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.k_proj.scales": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.k_proj.biases": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.q_proj.weight": "model-00067-of-00075.safetensors",
+    "model.layers.84.self_attn.q_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attn.q_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attention_res.proj_weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attention_res.norm_weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.mlp_res.proj_weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.mlp_res.norm_weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.e_score_correction_bias": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attn.q_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attn.k_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.self_attn.v_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.up_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.up_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.down_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.84.block_sparse_moe.switch_mlp.down_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_norm.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_up_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_up_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_up_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_down_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_down_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.routed_expert_down_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.up_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.up_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.up_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.down_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.down_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.down_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.gate_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.gate_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.shared_experts.gate_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.gate.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.gate.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.gate.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.post_attention_layernorm.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.input_layernorm.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.g_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.g_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.g_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.dt_bias": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_b_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_b_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_b_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_a_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_a_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.f_a_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.A_log": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.o_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.o_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.o_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.o_norm.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.b_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.b_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.b_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.v_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.v_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.v_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.k_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.k_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.k_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.q_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.q_proj.scales": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.q_proj.biases": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attention_res.proj_weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attention_res.norm_weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.mlp_res.proj_weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.mlp_res.norm_weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.e_score_correction_bias": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.q_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.k_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.self_attn.v_conv.conv.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00068-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.up_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.up_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.down_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.85.block_sparse_moe.switch_mlp.down_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_norm.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_up_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_up_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_up_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_down_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_down_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.routed_expert_down_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.up_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.up_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.up_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.down_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.down_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.down_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.gate_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.gate_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.shared_experts.gate_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.gate.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.gate.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.gate.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.post_attention_layernorm.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.input_layernorm.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.g_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.g_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.g_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.dt_bias": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_b_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_b_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_b_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_a_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_a_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.f_a_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.A_log": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.o_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.o_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.o_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.o_norm.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.b_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.b_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.b_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.v_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.v_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.v_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.k_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.k_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.k_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.q_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.q_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.q_proj.biases": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attention_res.proj_weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attention_res.norm_weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.mlp_res.proj_weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.mlp_res.norm_weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.e_score_correction_bias": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.q_conv.conv.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.k_conv.conv.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.self_attn.v_conv.conv.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.up_proj.weight": "model-00069-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.up_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.down_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.86.block_sparse_moe.switch_mlp.down_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.kv_a_proj_with_mqa.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.kv_a_proj_with_mqa.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.kv_a_proj_with_mqa.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_b_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_b_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_b_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_norm.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_up_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_up_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_up_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_down_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_down_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.routed_expert_down_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.up_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.up_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.up_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.down_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.down_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.down_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.gate_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.gate_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.shared_experts.gate_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.gate.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.gate.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.gate.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.post_attention_layernorm.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.input_layernorm.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.g_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.g_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.g_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.o_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.o_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.o_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.kv_a_layernorm.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_a_layernorm.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_a_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_a_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.q_a_proj.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attention_res.proj_weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attention_res.norm_weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.mlp_res.proj_weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.mlp_res.norm_weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.e_score_correction_bias": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.embed_q.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.embed_q.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.embed_q.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.unembed_out.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.unembed_out.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.self_attn.unembed_out.biases": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.up_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.up_proj.scales": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.down_proj.weight": "model-00070-of-00075.safetensors",
+    "model.layers.87.block_sparse_moe.switch_mlp.down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_norm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_up_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_up_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_up_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_down_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.routed_expert_down_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.up_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.up_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.up_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.down_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.down_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.gate_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.gate_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.shared_experts.gate_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.gate.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.gate.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.gate.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.post_attention_layernorm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.input_layernorm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.g_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.g_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.g_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.dt_bias": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_b_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_b_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_b_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_a_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_a_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.f_a_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.A_log": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.o_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.o_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.o_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.o_norm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.b_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.b_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.b_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.v_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.v_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.v_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.k_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.k_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.k_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.q_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.q_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.q_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attention_res.proj_weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attention_res.norm_weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.mlp_res.proj_weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.mlp_res.norm_weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.e_score_correction_bias": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.q_conv.conv.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.k_conv.conv.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.self_attn.v_conv.conv.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.up_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.up_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.down_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.88.block_sparse_moe.switch_mlp.down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_norm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_up_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_up_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_up_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_down_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.routed_expert_down_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.up_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.up_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.up_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.down_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.down_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.down_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.gate_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.gate_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.shared_experts.gate_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.gate.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.gate.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.gate.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.post_attention_layernorm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.input_layernorm.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.g_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.g_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.g_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.dt_bias": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_b_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_b_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_b_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_a_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_a_proj.scales": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.f_a_proj.biases": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.A_log": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.o_proj.weight": "model-00071-of-00075.safetensors",
+    "model.layers.89.self_attn.o_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.o_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.o_norm.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.b_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.b_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.b_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.v_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.v_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.v_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.k_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.k_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.k_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.q_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.q_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.q_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attention_res.proj_weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attention_res.norm_weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.mlp_res.proj_weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.mlp_res.norm_weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.e_score_correction_bias": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.q_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.k_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.self_attn.v_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.up_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.up_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.down_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.89.block_sparse_moe.switch_mlp.down_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_norm.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_up_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_up_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_up_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_down_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_down_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.routed_expert_down_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.up_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.up_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.up_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.down_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.down_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.down_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.gate_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.gate_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.shared_experts.gate_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.gate.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.gate.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.gate.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.post_attention_layernorm.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.input_layernorm.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.g_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.g_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.g_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.dt_bias": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_b_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_b_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_b_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_a_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_a_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.f_a_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.A_log": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.o_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.o_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.o_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.o_norm.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.b_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.b_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.b_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.v_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.v_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.v_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.k_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.k_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.k_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.q_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.q_proj.scales": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.q_proj.biases": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attention_res.proj_weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attention_res.norm_weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.mlp_res.proj_weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.mlp_res.norm_weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.e_score_correction_bias": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.q_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.k_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.self_attn.v_conv.conv.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00072-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.up_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.up_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.down_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.90.block_sparse_moe.switch_mlp.down_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.kv_a_proj_with_mqa.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.kv_a_proj_with_mqa.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.kv_a_proj_with_mqa.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_b_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_b_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_b_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_norm.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_up_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_up_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_up_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_down_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_down_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.routed_expert_down_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.up_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.up_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.up_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.down_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.down_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.down_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.gate_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.gate_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.shared_experts.gate_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.gate.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.gate.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.gate.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.post_attention_layernorm.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.input_layernorm.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.g_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.g_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.g_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.o_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.o_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.o_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.kv_a_layernorm.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_a_layernorm.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_a_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_a_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.q_a_proj.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attention_res.proj_weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attention_res.norm_weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.mlp_res.proj_weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.mlp_res.norm_weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.e_score_correction_bias": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.embed_q.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.embed_q.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.embed_q.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.unembed_out.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.unembed_out.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.self_attn.unembed_out.biases": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.up_proj.weight": "model-00073-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.up_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.down_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.91.block_sparse_moe.switch_mlp.down_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.kv_a_proj_with_mqa.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.kv_a_proj_with_mqa.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.kv_a_proj_with_mqa.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_b_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_b_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_b_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_norm.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_up_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_up_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_up_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_down_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_down_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.routed_expert_down_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.up_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.up_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.up_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.down_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.down_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.down_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.gate_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.gate_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.shared_experts.gate_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.gate.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.gate.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.gate.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.post_attention_layernorm.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.input_layernorm.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.g_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.g_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.g_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.o_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.o_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.o_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.kv_a_layernorm.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_a_layernorm.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_a_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_a_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.q_a_proj.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attention_res.proj_weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attention_res.norm_weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.mlp_res.proj_weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.mlp_res.norm_weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.e_score_correction_bias": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.embed_q.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.embed_q.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.embed_q.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.unembed_out.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.unembed_out.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.self_attn.unembed_out.biases": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.gate_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.gate_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.up_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.up_proj.scales": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.down_proj.weight": "model-00074-of-00075.safetensors",
+    "model.layers.92.block_sparse_moe.switch_mlp.down_proj.scales": "model-00075-of-00075.safetensors",
+    "mm_projector.post_norm.weight": "model-00075-of-00075.safetensors",
+    "mm_projector.proj.0.weight": "model-00075-of-00075.safetensors",
+    "mm_projector.proj.2.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.0.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.1.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.10.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.11.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.12.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.13.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.14.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.15.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.16.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.17.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.18.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.19.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.2.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.20.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.21.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.22.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.23.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.24.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.25.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.26.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.3.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.4.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.5.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.6.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.7.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.8.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.mlp.fc0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.mlp.fc1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.norm0.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.norm1.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.wo.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.blocks.9.wqkv.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.encoder.final_layernorm.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.patch_embed.pos_emb.weight": "model-00075-of-00075.safetensors",
+    "vision_tower.patch_embed.proj.weight": "model-00075-of-00075.safetensors",
+    "model.norm.weight": "model-00075-of-00075.safetensors",
+    "model.output_attn_res.proj_weight": "model-00075-of-00075.safetensors",
+    "model.output_attn_res.norm_weight": "model-00075-of-00075.safetensors",
+    "lm_head.weight": "model-00075-of-00075.safetensors",
+    "lm_head.scales": "model-00075-of-00075.safetensors",
+    "lm_head.biases": "model-00075-of-00075.safetensors"
+  }
+}
\ No newline at end of file