{
    "architectures": [
        "Lfm2MoeForCausalLM"
    ],
    "auto_map": {
        "AutoConfig": "configuration_lfm2_moe.Lfm2MoeConfig",
        "AutoModelForCausalLM": "modeling_lfm2_moe.Lfm2MoeForCausalLM"
    },
    "bos_token_id": 1,
    "conv_L_cache": 3,
    "conv_bias": false,
    "dtype": "bfloat16",
    "eos_token_id": 7,
    "hidden_size": 2048,
    "intermediate_size": 7168,
    "layer_types": [
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv",
        "full_attention",
        "conv",
        "conv"
    ],
    "max_position_embeddings": 128000,
    "model_type": "lfm2_moe",
    "moe_intermediate_size": 1792,
    "norm_eps": 1e-05,
    "norm_topk_prob": true,
    "num_attention_heads": 32,
    "num_dense_layers": 2,
    "num_experts": 32,
    "num_experts_per_tok": 4,
    "num_hidden_layers": 24,
    "num_key_value_heads": 8,
    "pad_token_id": 0,
    "quantization": {
        "group_size": 32,
        "bits": 8,
        "mode": "affine",
        "model.embed_tokens": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.2.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.3.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.6.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.7.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.10.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.11.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.14.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.15.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.18.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.19.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.21.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.22.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        }
    },
    "quantization_config": {
        "group_size": 32,
        "bits": 8,
        "mode": "affine",
        "model.embed_tokens": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.0.feed_forward.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.1.feed_forward.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.2.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.2.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.3.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.3.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.4.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.5.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.6.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.6.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.7.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.7.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.8.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.9.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.10.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.10.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.11.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.11.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.12.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.13.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.14.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.14.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.15.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.15.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.16.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.17.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.18.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.18.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.19.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.19.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.20.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.21.self_attn.q_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.k_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.v_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.self_attn.out_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.gate": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.21.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 8
        },
        "model.layers.22.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.22.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.conv.in_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.conv.out_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.gate": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.gate_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.up_proj": {
            "group_size": 32,
            "bits": 6
        },
        "model.layers.23.feed_forward.switch_mlp.down_proj": {
            "group_size": 32,
            "bits": 6
        }
    },
    "rope_theta": 1000000.0,
    "routed_scaling_factor": 1.0,
    "transformers_version": "4.56.1",
    "use_cache": true,
    "use_expert_bias": true,
    "vocab_size": 65536
}