fix: rename weight keys for vLLM Qwen3.5 compatibility (model.X -> model.language_model.X), fix quant_method and tokenizer_class

config.json

@@ -105,7 +105,7 @@
   "vision_end_token_id": 248054,
   "vision_start_token_id": 248053,
   "quantization_config": {
-    "quant_method": "polar",
+    "quant_method": "polarengine",
     "weight_bits": 5,
     "block_size": 128
   }

model-00000-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96e2255bd1311187ea4e58bef4de88d9ba9490a03f254d7abf2f9c79b5430c37
-size 5367777040
+oid sha256:5f55fd14854e6c20ff355c7c303c00babed3cd9498dbae8602ab3bf5a2abd7cf
+size 5367783760

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98ee9abed728472ec647df269266033872e153c59885e936ab2caeabe2d77bdc
-size 3727104936
+oid sha256:875d126f89aed4ca237dacb6c5d07e4358926c2d8d1f27a59ad24bef7ed6b63c
+size 3727112080

polar_config.json

@@ -5,7 +5,7 @@
   "quantization": "PolarQuant Q5 (Lloyd-Max + Hadamard)",
   "bits": 5,
   "layers": {
-    "model.embed_tokens": {
+    "model.language_model.embed_tokens": {
       "in_features": 4096,
       "out_features": 248320,
       "in_f_padded": 4096,
@@ -13,7 +13,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.linear_attn.out_proj": {
+    "model.language_model.layers.0.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -21,7 +21,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.0.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -29,7 +29,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.linear_attn.in_proj_z": {
+    "model.language_model.layers.0.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -37,7 +37,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.linear_attn.in_proj_b": {
+    "model.language_model.layers.0.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -45,7 +45,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.linear_attn.in_proj_a": {
+    "model.language_model.layers.0.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -53,7 +53,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.mlp.gate_proj": {
+    "model.language_model.layers.0.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -61,7 +61,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.mlp.up_proj": {
+    "model.language_model.layers.0.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -69,7 +69,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.0.mlp.down_proj": {
+    "model.language_model.layers.0.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -77,7 +77,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.linear_attn.out_proj": {
+    "model.language_model.layers.1.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -85,7 +85,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.1.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -93,7 +93,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.linear_attn.in_proj_z": {
+    "model.language_model.layers.1.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -101,7 +101,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.linear_attn.in_proj_b": {
+    "model.language_model.layers.1.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -109,7 +109,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.linear_attn.in_proj_a": {
+    "model.language_model.layers.1.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -117,7 +117,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.mlp.gate_proj": {
+    "model.language_model.layers.1.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -125,7 +125,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.mlp.up_proj": {
+    "model.language_model.layers.1.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -133,7 +133,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.1.mlp.down_proj": {
+    "model.language_model.layers.1.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -141,7 +141,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.linear_attn.out_proj": {
+    "model.language_model.layers.2.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -149,7 +149,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.2.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -157,7 +157,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.linear_attn.in_proj_z": {
+    "model.language_model.layers.2.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -165,7 +165,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.linear_attn.in_proj_b": {
+    "model.language_model.layers.2.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -173,7 +173,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.linear_attn.in_proj_a": {
+    "model.language_model.layers.2.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -181,7 +181,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.mlp.gate_proj": {
+    "model.language_model.layers.2.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -189,7 +189,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.mlp.up_proj": {
+    "model.language_model.layers.2.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -197,7 +197,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.2.mlp.down_proj": {
+    "model.language_model.layers.2.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -205,7 +205,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.self_attn.q_proj": {
+    "model.language_model.layers.3.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -213,7 +213,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.self_attn.k_proj": {
+    "model.language_model.layers.3.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -221,7 +221,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.self_attn.v_proj": {
+    "model.language_model.layers.3.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -229,7 +229,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.self_attn.o_proj": {
+    "model.language_model.layers.3.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -237,7 +237,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.mlp.gate_proj": {
+    "model.language_model.layers.3.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -245,7 +245,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.mlp.up_proj": {
+    "model.language_model.layers.3.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -253,7 +253,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.3.mlp.down_proj": {
+    "model.language_model.layers.3.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -261,7 +261,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.linear_attn.out_proj": {
+    "model.language_model.layers.4.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -269,7 +269,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.4.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -277,7 +277,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.linear_attn.in_proj_z": {
+    "model.language_model.layers.4.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -285,7 +285,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.linear_attn.in_proj_b": {
+    "model.language_model.layers.4.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -293,7 +293,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.linear_attn.in_proj_a": {
+    "model.language_model.layers.4.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -301,7 +301,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.mlp.gate_proj": {
+    "model.language_model.layers.4.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -309,7 +309,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.mlp.up_proj": {
+    "model.language_model.layers.4.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -317,7 +317,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.4.mlp.down_proj": {
+    "model.language_model.layers.4.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -325,7 +325,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.linear_attn.out_proj": {
+    "model.language_model.layers.5.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -333,7 +333,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.5.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -341,7 +341,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.linear_attn.in_proj_z": {
+    "model.language_model.layers.5.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -349,7 +349,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.linear_attn.in_proj_b": {
+    "model.language_model.layers.5.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -357,7 +357,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.linear_attn.in_proj_a": {
+    "model.language_model.layers.5.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -365,7 +365,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.mlp.gate_proj": {
+    "model.language_model.layers.5.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -373,7 +373,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.mlp.up_proj": {
+    "model.language_model.layers.5.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -381,7 +381,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.5.mlp.down_proj": {
+    "model.language_model.layers.5.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -389,7 +389,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.linear_attn.out_proj": {
+    "model.language_model.layers.6.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -397,7 +397,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.6.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -405,7 +405,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.linear_attn.in_proj_z": {
+    "model.language_model.layers.6.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -413,7 +413,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.linear_attn.in_proj_b": {
+    "model.language_model.layers.6.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -421,7 +421,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.linear_attn.in_proj_a": {
+    "model.language_model.layers.6.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -429,7 +429,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.mlp.gate_proj": {
+    "model.language_model.layers.6.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -437,7 +437,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.mlp.up_proj": {
+    "model.language_model.layers.6.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -445,7 +445,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.6.mlp.down_proj": {
+    "model.language_model.layers.6.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -453,7 +453,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.self_attn.q_proj": {
+    "model.language_model.layers.7.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -461,7 +461,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.self_attn.k_proj": {
+    "model.language_model.layers.7.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -469,7 +469,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.self_attn.v_proj": {
+    "model.language_model.layers.7.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -477,7 +477,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.self_attn.o_proj": {
+    "model.language_model.layers.7.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -485,7 +485,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.mlp.gate_proj": {
+    "model.language_model.layers.7.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -493,7 +493,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.mlp.up_proj": {
+    "model.language_model.layers.7.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -501,7 +501,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.7.mlp.down_proj": {
+    "model.language_model.layers.7.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -509,7 +509,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.linear_attn.out_proj": {
+    "model.language_model.layers.8.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -517,7 +517,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.8.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -525,7 +525,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.linear_attn.in_proj_z": {
+    "model.language_model.layers.8.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -533,7 +533,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.linear_attn.in_proj_b": {
+    "model.language_model.layers.8.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -541,7 +541,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.linear_attn.in_proj_a": {
+    "model.language_model.layers.8.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -549,7 +549,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.mlp.gate_proj": {
+    "model.language_model.layers.8.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -557,7 +557,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.mlp.up_proj": {
+    "model.language_model.layers.8.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -565,7 +565,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.8.mlp.down_proj": {
+    "model.language_model.layers.8.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -573,7 +573,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.linear_attn.out_proj": {
+    "model.language_model.layers.9.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -581,7 +581,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.9.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -589,7 +589,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.linear_attn.in_proj_z": {
+    "model.language_model.layers.9.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -597,7 +597,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.linear_attn.in_proj_b": {
+    "model.language_model.layers.9.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -605,7 +605,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.linear_attn.in_proj_a": {
+    "model.language_model.layers.9.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -613,7 +613,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.mlp.gate_proj": {
+    "model.language_model.layers.9.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -621,7 +621,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.mlp.up_proj": {
+    "model.language_model.layers.9.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -629,7 +629,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.9.mlp.down_proj": {
+    "model.language_model.layers.9.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -637,7 +637,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.linear_attn.out_proj": {
+    "model.language_model.layers.10.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -645,7 +645,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.10.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -653,7 +653,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.linear_attn.in_proj_z": {
+    "model.language_model.layers.10.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -661,7 +661,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.linear_attn.in_proj_b": {
+    "model.language_model.layers.10.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -669,7 +669,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.linear_attn.in_proj_a": {
+    "model.language_model.layers.10.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -677,7 +677,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.mlp.gate_proj": {
+    "model.language_model.layers.10.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -685,7 +685,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.mlp.up_proj": {
+    "model.language_model.layers.10.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -693,7 +693,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.10.mlp.down_proj": {
+    "model.language_model.layers.10.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -701,7 +701,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.self_attn.q_proj": {
+    "model.language_model.layers.11.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -709,7 +709,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.self_attn.k_proj": {
+    "model.language_model.layers.11.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -717,7 +717,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.self_attn.v_proj": {
+    "model.language_model.layers.11.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -725,7 +725,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.self_attn.o_proj": {
+    "model.language_model.layers.11.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -733,7 +733,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.mlp.gate_proj": {
+    "model.language_model.layers.11.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -741,7 +741,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.mlp.up_proj": {
+    "model.language_model.layers.11.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -749,7 +749,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.11.mlp.down_proj": {
+    "model.language_model.layers.11.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -757,7 +757,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.linear_attn.out_proj": {
+    "model.language_model.layers.12.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -765,7 +765,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.12.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -773,7 +773,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.linear_attn.in_proj_z": {
+    "model.language_model.layers.12.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -781,7 +781,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.linear_attn.in_proj_b": {
+    "model.language_model.layers.12.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -789,7 +789,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.linear_attn.in_proj_a": {
+    "model.language_model.layers.12.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -797,7 +797,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.mlp.gate_proj": {
+    "model.language_model.layers.12.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -805,7 +805,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.mlp.up_proj": {
+    "model.language_model.layers.12.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -813,7 +813,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.12.mlp.down_proj": {
+    "model.language_model.layers.12.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -821,7 +821,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.linear_attn.out_proj": {
+    "model.language_model.layers.13.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -829,7 +829,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.13.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -837,7 +837,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.linear_attn.in_proj_z": {
+    "model.language_model.layers.13.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -845,7 +845,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.linear_attn.in_proj_b": {
+    "model.language_model.layers.13.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -853,7 +853,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.linear_attn.in_proj_a": {
+    "model.language_model.layers.13.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -861,7 +861,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.mlp.gate_proj": {
+    "model.language_model.layers.13.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -869,7 +869,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.mlp.up_proj": {
+    "model.language_model.layers.13.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -877,7 +877,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.13.mlp.down_proj": {
+    "model.language_model.layers.13.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -885,7 +885,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.linear_attn.out_proj": {
+    "model.language_model.layers.14.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -893,7 +893,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.14.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -901,7 +901,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.linear_attn.in_proj_z": {
+    "model.language_model.layers.14.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -909,7 +909,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.linear_attn.in_proj_b": {
+    "model.language_model.layers.14.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -917,7 +917,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.linear_attn.in_proj_a": {
+    "model.language_model.layers.14.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -925,7 +925,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.mlp.gate_proj": {
+    "model.language_model.layers.14.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -933,7 +933,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.mlp.up_proj": {
+    "model.language_model.layers.14.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -941,7 +941,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.14.mlp.down_proj": {
+    "model.language_model.layers.14.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -949,7 +949,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.self_attn.q_proj": {
+    "model.language_model.layers.15.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -957,7 +957,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.self_attn.k_proj": {
+    "model.language_model.layers.15.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -965,7 +965,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.self_attn.v_proj": {
+    "model.language_model.layers.15.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -973,7 +973,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.self_attn.o_proj": {
+    "model.language_model.layers.15.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -981,7 +981,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.mlp.gate_proj": {
+    "model.language_model.layers.15.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -989,7 +989,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.mlp.up_proj": {
+    "model.language_model.layers.15.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -997,7 +997,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.15.mlp.down_proj": {
+    "model.language_model.layers.15.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1005,7 +1005,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.linear_attn.out_proj": {
+    "model.language_model.layers.16.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1013,7 +1013,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.16.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1021,7 +1021,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.linear_attn.in_proj_z": {
+    "model.language_model.layers.16.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1029,7 +1029,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.linear_attn.in_proj_b": {
+    "model.language_model.layers.16.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1037,7 +1037,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.linear_attn.in_proj_a": {
+    "model.language_model.layers.16.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1045,7 +1045,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.mlp.gate_proj": {
+    "model.language_model.layers.16.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1053,7 +1053,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.mlp.up_proj": {
+    "model.language_model.layers.16.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1061,7 +1061,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.16.mlp.down_proj": {
+    "model.language_model.layers.16.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1069,7 +1069,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.linear_attn.out_proj": {
+    "model.language_model.layers.17.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1077,7 +1077,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.17.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1085,7 +1085,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.linear_attn.in_proj_z": {
+    "model.language_model.layers.17.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1093,7 +1093,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.linear_attn.in_proj_b": {
+    "model.language_model.layers.17.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1101,7 +1101,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.linear_attn.in_proj_a": {
+    "model.language_model.layers.17.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1109,7 +1109,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.mlp.gate_proj": {
+    "model.language_model.layers.17.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1117,7 +1117,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.mlp.up_proj": {
+    "model.language_model.layers.17.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1125,7 +1125,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.17.mlp.down_proj": {
+    "model.language_model.layers.17.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1133,7 +1133,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.linear_attn.out_proj": {
+    "model.language_model.layers.18.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1141,7 +1141,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.18.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1149,7 +1149,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.linear_attn.in_proj_z": {
+    "model.language_model.layers.18.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1157,7 +1157,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.linear_attn.in_proj_b": {
+    "model.language_model.layers.18.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1165,7 +1165,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.linear_attn.in_proj_a": {
+    "model.language_model.layers.18.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1173,7 +1173,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.mlp.gate_proj": {
+    "model.language_model.layers.18.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1181,7 +1181,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.mlp.up_proj": {
+    "model.language_model.layers.18.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1189,7 +1189,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.18.mlp.down_proj": {
+    "model.language_model.layers.18.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1197,7 +1197,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.self_attn.q_proj": {
+    "model.language_model.layers.19.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1205,7 +1205,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.self_attn.k_proj": {
+    "model.language_model.layers.19.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1213,7 +1213,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.self_attn.v_proj": {
+    "model.language_model.layers.19.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1221,7 +1221,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.self_attn.o_proj": {
+    "model.language_model.layers.19.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1229,7 +1229,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.mlp.gate_proj": {
+    "model.language_model.layers.19.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1237,7 +1237,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.mlp.up_proj": {
+    "model.language_model.layers.19.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1245,7 +1245,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.19.mlp.down_proj": {
+    "model.language_model.layers.19.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1253,7 +1253,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.linear_attn.out_proj": {
+    "model.language_model.layers.20.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1261,7 +1261,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.20.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1269,7 +1269,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.linear_attn.in_proj_z": {
+    "model.language_model.layers.20.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1277,7 +1277,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.linear_attn.in_proj_b": {
+    "model.language_model.layers.20.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1285,7 +1285,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.linear_attn.in_proj_a": {
+    "model.language_model.layers.20.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1293,7 +1293,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.mlp.gate_proj": {
+    "model.language_model.layers.20.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1301,7 +1301,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.mlp.up_proj": {
+    "model.language_model.layers.20.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1309,7 +1309,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.20.mlp.down_proj": {
+    "model.language_model.layers.20.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1317,7 +1317,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.linear_attn.out_proj": {
+    "model.language_model.layers.21.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1325,7 +1325,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.21.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1333,7 +1333,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.linear_attn.in_proj_z": {
+    "model.language_model.layers.21.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1341,7 +1341,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.linear_attn.in_proj_b": {
+    "model.language_model.layers.21.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1349,7 +1349,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.linear_attn.in_proj_a": {
+    "model.language_model.layers.21.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1357,7 +1357,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.mlp.gate_proj": {
+    "model.language_model.layers.21.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1365,7 +1365,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.mlp.up_proj": {
+    "model.language_model.layers.21.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1373,7 +1373,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.21.mlp.down_proj": {
+    "model.language_model.layers.21.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1381,7 +1381,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.linear_attn.out_proj": {
+    "model.language_model.layers.22.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1389,7 +1389,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.22.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1397,7 +1397,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.linear_attn.in_proj_z": {
+    "model.language_model.layers.22.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1405,7 +1405,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.linear_attn.in_proj_b": {
+    "model.language_model.layers.22.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1413,7 +1413,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.linear_attn.in_proj_a": {
+    "model.language_model.layers.22.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1421,7 +1421,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.mlp.gate_proj": {
+    "model.language_model.layers.22.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1429,7 +1429,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.mlp.up_proj": {
+    "model.language_model.layers.22.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1437,7 +1437,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.22.mlp.down_proj": {
+    "model.language_model.layers.22.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1445,7 +1445,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.self_attn.q_proj": {
+    "model.language_model.layers.23.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1453,7 +1453,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.self_attn.k_proj": {
+    "model.language_model.layers.23.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1461,7 +1461,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.self_attn.v_proj": {
+    "model.language_model.layers.23.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1469,7 +1469,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.self_attn.o_proj": {
+    "model.language_model.layers.23.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1477,7 +1477,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.mlp.gate_proj": {
+    "model.language_model.layers.23.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1485,7 +1485,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.mlp.up_proj": {
+    "model.language_model.layers.23.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1493,7 +1493,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.23.mlp.down_proj": {
+    "model.language_model.layers.23.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1501,7 +1501,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.linear_attn.out_proj": {
+    "model.language_model.layers.24.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1509,7 +1509,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.24.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1517,7 +1517,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.linear_attn.in_proj_z": {
+    "model.language_model.layers.24.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1525,7 +1525,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.linear_attn.in_proj_b": {
+    "model.language_model.layers.24.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1533,7 +1533,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.linear_attn.in_proj_a": {
+    "model.language_model.layers.24.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1541,7 +1541,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.mlp.gate_proj": {
+    "model.language_model.layers.24.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1549,7 +1549,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.mlp.up_proj": {
+    "model.language_model.layers.24.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1557,7 +1557,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.24.mlp.down_proj": {
+    "model.language_model.layers.24.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1565,7 +1565,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.linear_attn.out_proj": {
+    "model.language_model.layers.25.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1573,7 +1573,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.25.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1581,7 +1581,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.linear_attn.in_proj_z": {
+    "model.language_model.layers.25.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1589,7 +1589,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.linear_attn.in_proj_b": {
+    "model.language_model.layers.25.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1597,7 +1597,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.linear_attn.in_proj_a": {
+    "model.language_model.layers.25.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1605,7 +1605,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.mlp.gate_proj": {
+    "model.language_model.layers.25.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1613,7 +1613,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.mlp.up_proj": {
+    "model.language_model.layers.25.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1621,7 +1621,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.25.mlp.down_proj": {
+    "model.language_model.layers.25.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1629,7 +1629,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.linear_attn.out_proj": {
+    "model.language_model.layers.26.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1637,7 +1637,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.26.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1645,7 +1645,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.linear_attn.in_proj_z": {
+    "model.language_model.layers.26.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1653,7 +1653,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.linear_attn.in_proj_b": {
+    "model.language_model.layers.26.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1661,7 +1661,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.linear_attn.in_proj_a": {
+    "model.language_model.layers.26.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1669,7 +1669,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.mlp.gate_proj": {
+    "model.language_model.layers.26.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1677,7 +1677,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.mlp.up_proj": {
+    "model.language_model.layers.26.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1685,7 +1685,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.26.mlp.down_proj": {
+    "model.language_model.layers.26.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1693,7 +1693,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.self_attn.q_proj": {
+    "model.language_model.layers.27.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1701,7 +1701,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.self_attn.k_proj": {
+    "model.language_model.layers.27.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1709,7 +1709,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.self_attn.v_proj": {
+    "model.language_model.layers.27.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1717,7 +1717,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.self_attn.o_proj": {
+    "model.language_model.layers.27.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1725,7 +1725,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.mlp.gate_proj": {
+    "model.language_model.layers.27.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1733,7 +1733,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.mlp.up_proj": {
+    "model.language_model.layers.27.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1741,7 +1741,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.27.mlp.down_proj": {
+    "model.language_model.layers.27.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1749,7 +1749,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.linear_attn.out_proj": {
+    "model.language_model.layers.28.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1757,7 +1757,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.28.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1765,7 +1765,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.linear_attn.in_proj_z": {
+    "model.language_model.layers.28.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1773,7 +1773,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.linear_attn.in_proj_b": {
+    "model.language_model.layers.28.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1781,7 +1781,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.linear_attn.in_proj_a": {
+    "model.language_model.layers.28.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1789,7 +1789,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.mlp.gate_proj": {
+    "model.language_model.layers.28.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1797,7 +1797,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.mlp.up_proj": {
+    "model.language_model.layers.28.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1805,7 +1805,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.28.mlp.down_proj": {
+    "model.language_model.layers.28.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1813,7 +1813,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.linear_attn.out_proj": {
+    "model.language_model.layers.29.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1821,7 +1821,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.29.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1829,7 +1829,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.linear_attn.in_proj_z": {
+    "model.language_model.layers.29.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1837,7 +1837,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.linear_attn.in_proj_b": {
+    "model.language_model.layers.29.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1845,7 +1845,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.linear_attn.in_proj_a": {
+    "model.language_model.layers.29.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1853,7 +1853,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.mlp.gate_proj": {
+    "model.language_model.layers.29.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1861,7 +1861,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.mlp.up_proj": {
+    "model.language_model.layers.29.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1869,7 +1869,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.29.mlp.down_proj": {
+    "model.language_model.layers.29.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1877,7 +1877,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.linear_attn.out_proj": {
+    "model.language_model.layers.30.linear_attn.out_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1885,7 +1885,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.linear_attn.in_proj_qkv": {
+    "model.language_model.layers.30.linear_attn.in_proj_qkv": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1893,7 +1893,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.linear_attn.in_proj_z": {
+    "model.language_model.layers.30.linear_attn.in_proj_z": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1901,7 +1901,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.linear_attn.in_proj_b": {
+    "model.language_model.layers.30.linear_attn.in_proj_b": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1909,7 +1909,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.linear_attn.in_proj_a": {
+    "model.language_model.layers.30.linear_attn.in_proj_a": {
       "in_features": 4096,
       "out_features": 32,
       "in_f_padded": 4096,
@@ -1917,7 +1917,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.mlp.gate_proj": {
+    "model.language_model.layers.30.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1925,7 +1925,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.mlp.up_proj": {
+    "model.language_model.layers.30.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1933,7 +1933,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.30.mlp.down_proj": {
+    "model.language_model.layers.30.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,
@@ -1941,7 +1941,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.self_attn.q_proj": {
+    "model.language_model.layers.31.self_attn.q_proj": {
       "in_features": 4096,
       "out_features": 8192,
       "in_f_padded": 4096,
@@ -1949,7 +1949,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.self_attn.k_proj": {
+    "model.language_model.layers.31.self_attn.k_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1957,7 +1957,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.self_attn.v_proj": {
+    "model.language_model.layers.31.self_attn.v_proj": {
       "in_features": 4096,
       "out_features": 1024,
       "in_f_padded": 4096,
@@ -1965,7 +1965,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.self_attn.o_proj": {
+    "model.language_model.layers.31.self_attn.o_proj": {
       "in_features": 4096,
       "out_features": 4096,
       "in_f_padded": 4096,
@@ -1973,7 +1973,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.mlp.gate_proj": {
+    "model.language_model.layers.31.mlp.gate_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1981,7 +1981,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.mlp.up_proj": {
+    "model.language_model.layers.31.mlp.up_proj": {
       "in_features": 4096,
       "out_features": 12288,
       "in_f_padded": 4096,
@@ -1989,7 +1989,7 @@
       "bits": 5,
       "block_size": 128
     },
-    "model.layers.31.mlp.down_proj": {
+    "model.language_model.layers.31.mlp.down_proj": {
       "in_features": 12288,
       "out_features": 4096,
       "in_f_padded": 12288,

tokenizer_config.json

@@ -23,9 +23,9 @@
   "pad_token": "<|endoftext|>",
   "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
   "split_special_tokens": false,
-  "tokenizer_class": "TokenizersBackend",
+  "tokenizer_class": "Qwen2TokenizerFast",
   "unk_token": null,
   "video_token": "<|video_pad|>",
   "vision_bos_token": "<|vision_start|>",
   "vision_eos_token": "<|vision_end|>"
-}
+}