Add files using upload-large-folder tool
Browse files- config.json +506 -371
- model-00001-of-00009.safetensors +3 -0
- model-00002-of-00009.safetensors +3 -0
- model-00003-of-00009.safetensors +3 -0
- model-00004-of-00009.safetensors +3 -0
- model-00005-of-00009.safetensors +3 -0
- model-00006-of-00009.safetensors +3 -0
- model-00007-of-00009.safetensors +3 -0
- model-00008-of-00009.safetensors +3 -0
- model-00009-of-00009.safetensors +3 -0
- model.safetensors.index.json +2 -2
- old_config.json +647 -0
- recipe.yaml +1 -2
config.json
CHANGED
|
@@ -4,6 +4,511 @@
|
|
| 4 |
],
|
| 5 |
"image_token_id": 248056,
|
| 6 |
"model_type": "qwen3_5_moe",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
"text_config": {
|
| 8 |
"attention_bias": false,
|
| 9 |
"attention_dropout": 0.0,
|
|
@@ -131,375 +636,5 @@
|
|
| 131 |
"temporal_patch_size": 2
|
| 132 |
},
|
| 133 |
"vision_end_token_id": 248054,
|
| 134 |
-
"vision_start_token_id": 248053
|
| 135 |
-
"quantization_config": {
|
| 136 |
-
"config_groups": {
|
| 137 |
-
"group_0": {
|
| 138 |
-
"format": "float-quantized",
|
| 139 |
-
"input_activations": {
|
| 140 |
-
"actorder": null,
|
| 141 |
-
"block_structure": null,
|
| 142 |
-
"dynamic": true,
|
| 143 |
-
"group_size": null,
|
| 144 |
-
"num_bits": 8,
|
| 145 |
-
"observer": null,
|
| 146 |
-
"observer_kwargs": {},
|
| 147 |
-
"scale_dtype": null,
|
| 148 |
-
"strategy": "token",
|
| 149 |
-
"symmetric": true,
|
| 150 |
-
"type": "float",
|
| 151 |
-
"zp_dtype": null
|
| 152 |
-
},
|
| 153 |
-
"output_activations": null,
|
| 154 |
-
"targets": [
|
| 155 |
-
"Linear"
|
| 156 |
-
],
|
| 157 |
-
"weights": {
|
| 158 |
-
"actorder": null,
|
| 159 |
-
"block_structure": null,
|
| 160 |
-
"dynamic": false,
|
| 161 |
-
"group_size": null,
|
| 162 |
-
"num_bits": 8,
|
| 163 |
-
"observer": "memoryless_minmax",
|
| 164 |
-
"observer_kwargs": {},
|
| 165 |
-
"scale_dtype": null,
|
| 166 |
-
"strategy": "channel",
|
| 167 |
-
"symmetric": true,
|
| 168 |
-
"type": "float",
|
| 169 |
-
"zp_dtype": null
|
| 170 |
-
}
|
| 171 |
-
}
|
| 172 |
-
},
|
| 173 |
-
"format": "float-quantized",
|
| 174 |
-
"global_compression_ratio": null,
|
| 175 |
-
"ignore": [
|
| 176 |
-
"model.visual.blocks.0.attn.qkv",
|
| 177 |
-
"model.visual.blocks.0.attn.proj",
|
| 178 |
-
"model.visual.blocks.0.mlp.linear_fc1",
|
| 179 |
-
"model.visual.blocks.0.mlp.linear_fc2",
|
| 180 |
-
"model.visual.blocks.1.attn.qkv",
|
| 181 |
-
"model.visual.blocks.1.attn.proj",
|
| 182 |
-
"model.visual.blocks.1.mlp.linear_fc1",
|
| 183 |
-
"model.visual.blocks.1.mlp.linear_fc2",
|
| 184 |
-
"model.visual.blocks.2.attn.qkv",
|
| 185 |
-
"model.visual.blocks.2.attn.proj",
|
| 186 |
-
"model.visual.blocks.2.mlp.linear_fc1",
|
| 187 |
-
"model.visual.blocks.2.mlp.linear_fc2",
|
| 188 |
-
"model.visual.blocks.3.attn.qkv",
|
| 189 |
-
"model.visual.blocks.3.attn.proj",
|
| 190 |
-
"model.visual.blocks.3.mlp.linear_fc1",
|
| 191 |
-
"model.visual.blocks.3.mlp.linear_fc2",
|
| 192 |
-
"model.visual.blocks.4.attn.qkv",
|
| 193 |
-
"model.visual.blocks.4.attn.proj",
|
| 194 |
-
"model.visual.blocks.4.mlp.linear_fc1",
|
| 195 |
-
"model.visual.blocks.4.mlp.linear_fc2",
|
| 196 |
-
"model.visual.blocks.5.attn.qkv",
|
| 197 |
-
"model.visual.blocks.5.attn.proj",
|
| 198 |
-
"model.visual.blocks.5.mlp.linear_fc1",
|
| 199 |
-
"model.visual.blocks.5.mlp.linear_fc2",
|
| 200 |
-
"model.visual.blocks.6.attn.qkv",
|
| 201 |
-
"model.visual.blocks.6.attn.proj",
|
| 202 |
-
"model.visual.blocks.6.mlp.linear_fc1",
|
| 203 |
-
"model.visual.blocks.6.mlp.linear_fc2",
|
| 204 |
-
"model.visual.blocks.7.attn.qkv",
|
| 205 |
-
"model.visual.blocks.7.attn.proj",
|
| 206 |
-
"model.visual.blocks.7.mlp.linear_fc1",
|
| 207 |
-
"model.visual.blocks.7.mlp.linear_fc2",
|
| 208 |
-
"model.visual.blocks.8.attn.qkv",
|
| 209 |
-
"model.visual.blocks.8.attn.proj",
|
| 210 |
-
"model.visual.blocks.8.mlp.linear_fc1",
|
| 211 |
-
"model.visual.blocks.8.mlp.linear_fc2",
|
| 212 |
-
"model.visual.blocks.9.attn.qkv",
|
| 213 |
-
"model.visual.blocks.9.attn.proj",
|
| 214 |
-
"model.visual.blocks.9.mlp.linear_fc1",
|
| 215 |
-
"model.visual.blocks.9.mlp.linear_fc2",
|
| 216 |
-
"model.visual.blocks.10.attn.qkv",
|
| 217 |
-
"model.visual.blocks.10.attn.proj",
|
| 218 |
-
"model.visual.blocks.10.mlp.linear_fc1",
|
| 219 |
-
"model.visual.blocks.10.mlp.linear_fc2",
|
| 220 |
-
"model.visual.blocks.11.attn.qkv",
|
| 221 |
-
"model.visual.blocks.11.attn.proj",
|
| 222 |
-
"model.visual.blocks.11.mlp.linear_fc1",
|
| 223 |
-
"model.visual.blocks.11.mlp.linear_fc2",
|
| 224 |
-
"model.visual.blocks.12.attn.qkv",
|
| 225 |
-
"model.visual.blocks.12.attn.proj",
|
| 226 |
-
"model.visual.blocks.12.mlp.linear_fc1",
|
| 227 |
-
"model.visual.blocks.12.mlp.linear_fc2",
|
| 228 |
-
"model.visual.blocks.13.attn.qkv",
|
| 229 |
-
"model.visual.blocks.13.attn.proj",
|
| 230 |
-
"model.visual.blocks.13.mlp.linear_fc1",
|
| 231 |
-
"model.visual.blocks.13.mlp.linear_fc2",
|
| 232 |
-
"model.visual.blocks.14.attn.qkv",
|
| 233 |
-
"model.visual.blocks.14.attn.proj",
|
| 234 |
-
"model.visual.blocks.14.mlp.linear_fc1",
|
| 235 |
-
"model.visual.blocks.14.mlp.linear_fc2",
|
| 236 |
-
"model.visual.blocks.15.attn.qkv",
|
| 237 |
-
"model.visual.blocks.15.attn.proj",
|
| 238 |
-
"model.visual.blocks.15.mlp.linear_fc1",
|
| 239 |
-
"model.visual.blocks.15.mlp.linear_fc2",
|
| 240 |
-
"model.visual.blocks.16.attn.qkv",
|
| 241 |
-
"model.visual.blocks.16.attn.proj",
|
| 242 |
-
"model.visual.blocks.16.mlp.linear_fc1",
|
| 243 |
-
"model.visual.blocks.16.mlp.linear_fc2",
|
| 244 |
-
"model.visual.blocks.17.attn.qkv",
|
| 245 |
-
"model.visual.blocks.17.attn.proj",
|
| 246 |
-
"model.visual.blocks.17.mlp.linear_fc1",
|
| 247 |
-
"model.visual.blocks.17.mlp.linear_fc2",
|
| 248 |
-
"model.visual.blocks.18.attn.qkv",
|
| 249 |
-
"model.visual.blocks.18.attn.proj",
|
| 250 |
-
"model.visual.blocks.18.mlp.linear_fc1",
|
| 251 |
-
"model.visual.blocks.18.mlp.linear_fc2",
|
| 252 |
-
"model.visual.blocks.19.attn.qkv",
|
| 253 |
-
"model.visual.blocks.19.attn.proj",
|
| 254 |
-
"model.visual.blocks.19.mlp.linear_fc1",
|
| 255 |
-
"model.visual.blocks.19.mlp.linear_fc2",
|
| 256 |
-
"model.visual.blocks.20.attn.qkv",
|
| 257 |
-
"model.visual.blocks.20.attn.proj",
|
| 258 |
-
"model.visual.blocks.20.mlp.linear_fc1",
|
| 259 |
-
"model.visual.blocks.20.mlp.linear_fc2",
|
| 260 |
-
"model.visual.blocks.21.attn.qkv",
|
| 261 |
-
"model.visual.blocks.21.attn.proj",
|
| 262 |
-
"model.visual.blocks.21.mlp.linear_fc1",
|
| 263 |
-
"model.visual.blocks.21.mlp.linear_fc2",
|
| 264 |
-
"model.visual.blocks.22.attn.qkv",
|
| 265 |
-
"model.visual.blocks.22.attn.proj",
|
| 266 |
-
"model.visual.blocks.22.mlp.linear_fc1",
|
| 267 |
-
"model.visual.blocks.22.mlp.linear_fc2",
|
| 268 |
-
"model.visual.blocks.23.attn.qkv",
|
| 269 |
-
"model.visual.blocks.23.attn.proj",
|
| 270 |
-
"model.visual.blocks.23.mlp.linear_fc1",
|
| 271 |
-
"model.visual.blocks.23.mlp.linear_fc2",
|
| 272 |
-
"model.visual.blocks.24.attn.qkv",
|
| 273 |
-
"model.visual.blocks.24.attn.proj",
|
| 274 |
-
"model.visual.blocks.24.mlp.linear_fc1",
|
| 275 |
-
"model.visual.blocks.24.mlp.linear_fc2",
|
| 276 |
-
"model.visual.blocks.25.attn.qkv",
|
| 277 |
-
"model.visual.blocks.25.attn.proj",
|
| 278 |
-
"model.visual.blocks.25.mlp.linear_fc1",
|
| 279 |
-
"model.visual.blocks.25.mlp.linear_fc2",
|
| 280 |
-
"model.visual.blocks.26.attn.qkv",
|
| 281 |
-
"model.visual.blocks.26.attn.proj",
|
| 282 |
-
"model.visual.blocks.26.mlp.linear_fc1",
|
| 283 |
-
"model.visual.blocks.26.mlp.linear_fc2",
|
| 284 |
-
"model.visual.merger.linear_fc1",
|
| 285 |
-
"model.visual.merger.linear_fc2",
|
| 286 |
-
"model.language_model.layers.0.linear_attn.in_proj_b",
|
| 287 |
-
"model.language_model.layers.0.linear_attn.in_proj_a",
|
| 288 |
-
"model.language_model.layers.0.mlp.shared_expert_gate",
|
| 289 |
-
"model.language_model.layers.1.linear_attn.in_proj_b",
|
| 290 |
-
"model.language_model.layers.1.linear_attn.in_proj_a",
|
| 291 |
-
"model.language_model.layers.1.mlp.shared_expert_gate",
|
| 292 |
-
"model.language_model.layers.2.linear_attn.in_proj_b",
|
| 293 |
-
"model.language_model.layers.2.linear_attn.in_proj_a",
|
| 294 |
-
"model.language_model.layers.2.mlp.shared_expert_gate",
|
| 295 |
-
"model.language_model.layers.3.mlp.shared_expert_gate",
|
| 296 |
-
"model.language_model.layers.4.linear_attn.in_proj_b",
|
| 297 |
-
"model.language_model.layers.4.linear_attn.in_proj_a",
|
| 298 |
-
"model.language_model.layers.4.mlp.shared_expert_gate",
|
| 299 |
-
"model.language_model.layers.5.linear_attn.in_proj_b",
|
| 300 |
-
"model.language_model.layers.5.linear_attn.in_proj_a",
|
| 301 |
-
"model.language_model.layers.5.mlp.shared_expert_gate",
|
| 302 |
-
"model.language_model.layers.6.linear_attn.in_proj_b",
|
| 303 |
-
"model.language_model.layers.6.linear_attn.in_proj_a",
|
| 304 |
-
"model.language_model.layers.6.mlp.shared_expert_gate",
|
| 305 |
-
"model.language_model.layers.7.mlp.shared_expert_gate",
|
| 306 |
-
"model.language_model.layers.8.linear_attn.in_proj_b",
|
| 307 |
-
"model.language_model.layers.8.linear_attn.in_proj_a",
|
| 308 |
-
"model.language_model.layers.8.mlp.shared_expert_gate",
|
| 309 |
-
"model.language_model.layers.9.linear_attn.in_proj_b",
|
| 310 |
-
"model.language_model.layers.9.linear_attn.in_proj_a",
|
| 311 |
-
"model.language_model.layers.9.mlp.shared_expert_gate",
|
| 312 |
-
"model.language_model.layers.10.linear_attn.in_proj_b",
|
| 313 |
-
"model.language_model.layers.10.linear_attn.in_proj_a",
|
| 314 |
-
"model.language_model.layers.10.mlp.shared_expert_gate",
|
| 315 |
-
"model.language_model.layers.11.mlp.shared_expert_gate",
|
| 316 |
-
"model.language_model.layers.12.linear_attn.in_proj_b",
|
| 317 |
-
"model.language_model.layers.12.linear_attn.in_proj_a",
|
| 318 |
-
"model.language_model.layers.12.mlp.shared_expert_gate",
|
| 319 |
-
"model.language_model.layers.13.linear_attn.in_proj_b",
|
| 320 |
-
"model.language_model.layers.13.linear_attn.in_proj_a",
|
| 321 |
-
"model.language_model.layers.13.mlp.shared_expert_gate",
|
| 322 |
-
"model.language_model.layers.14.linear_attn.in_proj_b",
|
| 323 |
-
"model.language_model.layers.14.linear_attn.in_proj_a",
|
| 324 |
-
"model.language_model.layers.14.mlp.shared_expert_gate",
|
| 325 |
-
"model.language_model.layers.15.mlp.shared_expert_gate",
|
| 326 |
-
"model.language_model.layers.16.linear_attn.in_proj_b",
|
| 327 |
-
"model.language_model.layers.16.linear_attn.in_proj_a",
|
| 328 |
-
"model.language_model.layers.16.mlp.shared_expert_gate",
|
| 329 |
-
"model.language_model.layers.17.linear_attn.in_proj_b",
|
| 330 |
-
"model.language_model.layers.17.linear_attn.in_proj_a",
|
| 331 |
-
"model.language_model.layers.17.mlp.shared_expert_gate",
|
| 332 |
-
"model.language_model.layers.18.linear_attn.in_proj_b",
|
| 333 |
-
"model.language_model.layers.18.linear_attn.in_proj_a",
|
| 334 |
-
"model.language_model.layers.18.mlp.shared_expert_gate",
|
| 335 |
-
"model.language_model.layers.19.mlp.shared_expert_gate",
|
| 336 |
-
"model.language_model.layers.20.linear_attn.in_proj_b",
|
| 337 |
-
"model.language_model.layers.20.linear_attn.in_proj_a",
|
| 338 |
-
"model.language_model.layers.20.mlp.shared_expert_gate",
|
| 339 |
-
"model.language_model.layers.21.linear_attn.in_proj_b",
|
| 340 |
-
"model.language_model.layers.21.linear_attn.in_proj_a",
|
| 341 |
-
"model.language_model.layers.21.mlp.shared_expert_gate",
|
| 342 |
-
"model.language_model.layers.22.linear_attn.in_proj_b",
|
| 343 |
-
"model.language_model.layers.22.linear_attn.in_proj_a",
|
| 344 |
-
"model.language_model.layers.22.mlp.shared_expert_gate",
|
| 345 |
-
"model.language_model.layers.23.mlp.shared_expert_gate",
|
| 346 |
-
"model.language_model.layers.24.linear_attn.in_proj_b",
|
| 347 |
-
"model.language_model.layers.24.linear_attn.in_proj_a",
|
| 348 |
-
"model.language_model.layers.24.mlp.shared_expert_gate",
|
| 349 |
-
"model.language_model.layers.25.linear_attn.in_proj_b",
|
| 350 |
-
"model.language_model.layers.25.linear_attn.in_proj_a",
|
| 351 |
-
"model.language_model.layers.25.mlp.shared_expert_gate",
|
| 352 |
-
"model.language_model.layers.26.linear_attn.in_proj_b",
|
| 353 |
-
"model.language_model.layers.26.linear_attn.in_proj_a",
|
| 354 |
-
"model.language_model.layers.26.mlp.shared_expert_gate",
|
| 355 |
-
"model.language_model.layers.27.mlp.shared_expert_gate",
|
| 356 |
-
"model.language_model.layers.28.linear_attn.in_proj_b",
|
| 357 |
-
"model.language_model.layers.28.linear_attn.in_proj_a",
|
| 358 |
-
"model.language_model.layers.28.mlp.shared_expert_gate",
|
| 359 |
-
"model.language_model.layers.29.linear_attn.in_proj_b",
|
| 360 |
-
"model.language_model.layers.29.linear_attn.in_proj_a",
|
| 361 |
-
"model.language_model.layers.29.mlp.shared_expert_gate",
|
| 362 |
-
"model.language_model.layers.30.linear_attn.in_proj_b",
|
| 363 |
-
"model.language_model.layers.30.linear_attn.in_proj_a",
|
| 364 |
-
"model.language_model.layers.30.mlp.shared_expert_gate",
|
| 365 |
-
"model.language_model.layers.31.mlp.shared_expert_gate",
|
| 366 |
-
"model.language_model.layers.32.linear_attn.in_proj_b",
|
| 367 |
-
"model.language_model.layers.32.linear_attn.in_proj_a",
|
| 368 |
-
"model.language_model.layers.32.mlp.shared_expert_gate",
|
| 369 |
-
"model.language_model.layers.33.linear_attn.in_proj_b",
|
| 370 |
-
"model.language_model.layers.33.linear_attn.in_proj_a",
|
| 371 |
-
"model.language_model.layers.33.mlp.shared_expert_gate",
|
| 372 |
-
"model.language_model.layers.34.linear_attn.in_proj_b",
|
| 373 |
-
"model.language_model.layers.34.linear_attn.in_proj_a",
|
| 374 |
-
"model.language_model.layers.34.mlp.shared_expert_gate",
|
| 375 |
-
"model.language_model.layers.35.mlp.shared_expert_gate",
|
| 376 |
-
"model.language_model.layers.36.linear_attn.in_proj_b",
|
| 377 |
-
"model.language_model.layers.36.linear_attn.in_proj_a",
|
| 378 |
-
"model.language_model.layers.36.mlp.shared_expert_gate",
|
| 379 |
-
"model.language_model.layers.37.linear_attn.in_proj_b",
|
| 380 |
-
"model.language_model.layers.37.linear_attn.in_proj_a",
|
| 381 |
-
"model.language_model.layers.37.mlp.shared_expert_gate",
|
| 382 |
-
"model.language_model.layers.38.linear_attn.in_proj_b",
|
| 383 |
-
"model.language_model.layers.38.linear_attn.in_proj_a",
|
| 384 |
-
"model.language_model.layers.38.mlp.shared_expert_gate",
|
| 385 |
-
"model.language_model.layers.39.mlp.shared_expert_gate",
|
| 386 |
-
"model.language_model.layers.40.linear_attn.in_proj_b",
|
| 387 |
-
"model.language_model.layers.40.linear_attn.in_proj_a",
|
| 388 |
-
"model.language_model.layers.40.mlp.shared_expert_gate",
|
| 389 |
-
"model.language_model.layers.41.linear_attn.in_proj_b",
|
| 390 |
-
"model.language_model.layers.41.linear_attn.in_proj_a",
|
| 391 |
-
"model.language_model.layers.41.mlp.shared_expert_gate",
|
| 392 |
-
"model.language_model.layers.42.linear_attn.in_proj_b",
|
| 393 |
-
"model.language_model.layers.42.linear_attn.in_proj_a",
|
| 394 |
-
"model.language_model.layers.42.mlp.shared_expert_gate",
|
| 395 |
-
"model.language_model.layers.43.mlp.shared_expert_gate",
|
| 396 |
-
"model.language_model.layers.44.linear_attn.in_proj_b",
|
| 397 |
-
"model.language_model.layers.44.linear_attn.in_proj_a",
|
| 398 |
-
"model.language_model.layers.44.mlp.shared_expert_gate",
|
| 399 |
-
"model.language_model.layers.45.linear_attn.in_proj_b",
|
| 400 |
-
"model.language_model.layers.45.linear_attn.in_proj_a",
|
| 401 |
-
"model.language_model.layers.45.mlp.shared_expert_gate",
|
| 402 |
-
"model.language_model.layers.46.linear_attn.in_proj_b",
|
| 403 |
-
"model.language_model.layers.46.linear_attn.in_proj_a",
|
| 404 |
-
"model.language_model.layers.46.mlp.shared_expert_gate",
|
| 405 |
-
"model.language_model.layers.47.mlp.shared_expert_gate",
|
| 406 |
-
"model.language_model.layers.48.linear_attn.in_proj_b",
|
| 407 |
-
"model.language_model.layers.48.linear_attn.in_proj_a",
|
| 408 |
-
"model.language_model.layers.48.mlp.shared_expert_gate",
|
| 409 |
-
"model.language_model.layers.49.linear_attn.in_proj_b",
|
| 410 |
-
"model.language_model.layers.49.linear_attn.in_proj_a",
|
| 411 |
-
"model.language_model.layers.49.mlp.shared_expert_gate",
|
| 412 |
-
"model.language_model.layers.50.linear_attn.in_proj_b",
|
| 413 |
-
"model.language_model.layers.50.linear_attn.in_proj_a",
|
| 414 |
-
"model.language_model.layers.50.mlp.shared_expert_gate",
|
| 415 |
-
"model.language_model.layers.51.mlp.shared_expert_gate",
|
| 416 |
-
"model.language_model.layers.52.linear_attn.in_proj_b",
|
| 417 |
-
"model.language_model.layers.52.linear_attn.in_proj_a",
|
| 418 |
-
"model.language_model.layers.52.mlp.shared_expert_gate",
|
| 419 |
-
"model.language_model.layers.53.linear_attn.in_proj_b",
|
| 420 |
-
"model.language_model.layers.53.linear_attn.in_proj_a",
|
| 421 |
-
"model.language_model.layers.53.mlp.shared_expert_gate",
|
| 422 |
-
"model.language_model.layers.54.linear_attn.in_proj_b",
|
| 423 |
-
"model.language_model.layers.54.linear_attn.in_proj_a",
|
| 424 |
-
"model.language_model.layers.54.mlp.shared_expert_gate",
|
| 425 |
-
"model.language_model.layers.55.mlp.shared_expert_gate",
|
| 426 |
-
"model.language_model.layers.56.linear_attn.in_proj_b",
|
| 427 |
-
"model.language_model.layers.56.linear_attn.in_proj_a",
|
| 428 |
-
"model.language_model.layers.56.mlp.shared_expert_gate",
|
| 429 |
-
"model.language_model.layers.57.linear_attn.in_proj_b",
|
| 430 |
-
"model.language_model.layers.57.linear_attn.in_proj_a",
|
| 431 |
-
"model.language_model.layers.57.mlp.shared_expert_gate",
|
| 432 |
-
"model.language_model.layers.58.linear_attn.in_proj_b",
|
| 433 |
-
"model.language_model.layers.58.linear_attn.in_proj_a",
|
| 434 |
-
"model.language_model.layers.58.mlp.shared_expert_gate",
|
| 435 |
-
"model.language_model.layers.59.mlp.shared_expert_gate",
|
| 436 |
-
"model.language_model.layers.0.mlp.gate",
|
| 437 |
-
"model.language_model.layers.1.mlp.gate",
|
| 438 |
-
"model.language_model.layers.2.mlp.gate",
|
| 439 |
-
"model.language_model.layers.3.mlp.gate",
|
| 440 |
-
"model.language_model.layers.4.mlp.gate",
|
| 441 |
-
"model.language_model.layers.5.mlp.gate",
|
| 442 |
-
"model.language_model.layers.6.mlp.gate",
|
| 443 |
-
"model.language_model.layers.7.mlp.gate",
|
| 444 |
-
"model.language_model.layers.8.mlp.gate",
|
| 445 |
-
"model.language_model.layers.9.mlp.gate",
|
| 446 |
-
"model.language_model.layers.10.mlp.gate",
|
| 447 |
-
"model.language_model.layers.11.mlp.gate",
|
| 448 |
-
"model.language_model.layers.12.mlp.gate",
|
| 449 |
-
"model.language_model.layers.13.mlp.gate",
|
| 450 |
-
"model.language_model.layers.14.mlp.gate",
|
| 451 |
-
"model.language_model.layers.15.mlp.gate",
|
| 452 |
-
"model.language_model.layers.16.mlp.gate",
|
| 453 |
-
"model.language_model.layers.17.mlp.gate",
|
| 454 |
-
"model.language_model.layers.18.mlp.gate",
|
| 455 |
-
"model.language_model.layers.19.mlp.gate",
|
| 456 |
-
"model.language_model.layers.20.mlp.gate",
|
| 457 |
-
"model.language_model.layers.21.mlp.gate",
|
| 458 |
-
"model.language_model.layers.22.mlp.gate",
|
| 459 |
-
"model.language_model.layers.23.mlp.gate",
|
| 460 |
-
"model.language_model.layers.24.mlp.gate",
|
| 461 |
-
"model.language_model.layers.25.mlp.gate",
|
| 462 |
-
"model.language_model.layers.26.mlp.gate",
|
| 463 |
-
"model.language_model.layers.27.mlp.gate",
|
| 464 |
-
"model.language_model.layers.28.mlp.gate",
|
| 465 |
-
"model.language_model.layers.29.mlp.gate",
|
| 466 |
-
"model.language_model.layers.30.mlp.gate",
|
| 467 |
-
"model.language_model.layers.31.mlp.gate",
|
| 468 |
-
"model.language_model.layers.32.mlp.gate",
|
| 469 |
-
"model.language_model.layers.33.mlp.gate",
|
| 470 |
-
"model.language_model.layers.34.mlp.gate",
|
| 471 |
-
"model.language_model.layers.35.mlp.gate",
|
| 472 |
-
"model.language_model.layers.36.mlp.gate",
|
| 473 |
-
"model.language_model.layers.37.mlp.gate",
|
| 474 |
-
"model.language_model.layers.38.mlp.gate",
|
| 475 |
-
"model.language_model.layers.39.mlp.gate",
|
| 476 |
-
"model.language_model.layers.40.mlp.gate",
|
| 477 |
-
"model.language_model.layers.41.mlp.gate",
|
| 478 |
-
"model.language_model.layers.42.mlp.gate",
|
| 479 |
-
"model.language_model.layers.43.mlp.gate",
|
| 480 |
-
"model.language_model.layers.44.mlp.gate",
|
| 481 |
-
"model.language_model.layers.45.mlp.gate",
|
| 482 |
-
"model.language_model.layers.46.mlp.gate",
|
| 483 |
-
"model.language_model.layers.47.mlp.gate",
|
| 484 |
-
"model.language_model.layers.48.mlp.gate",
|
| 485 |
-
"model.language_model.layers.49.mlp.gate",
|
| 486 |
-
"model.language_model.layers.50.mlp.gate",
|
| 487 |
-
"model.language_model.layers.51.mlp.gate",
|
| 488 |
-
"model.language_model.layers.52.mlp.gate",
|
| 489 |
-
"model.language_model.layers.53.mlp.gate",
|
| 490 |
-
"model.language_model.layers.54.mlp.gate",
|
| 491 |
-
"model.language_model.layers.55.mlp.gate",
|
| 492 |
-
"model.language_model.layers.56.mlp.gate",
|
| 493 |
-
"model.language_model.layers.57.mlp.gate",
|
| 494 |
-
"model.language_model.layers.58.mlp.gate",
|
| 495 |
-
"model.language_model.layers.59.mlp.gate",
|
| 496 |
-
"lm_head"
|
| 497 |
-
],
|
| 498 |
-
"kv_cache_scheme": null,
|
| 499 |
-
"quant_method": "compressed-tensors",
|
| 500 |
-
"quantization_status": "compressed",
|
| 501 |
-
"sparsity_config": {},
|
| 502 |
-
"transform_config": {},
|
| 503 |
-
"version": "0.13.1.a20260212"
|
| 504 |
-
}
|
| 505 |
}
|
|
|
|
| 4 |
],
|
| 5 |
"image_token_id": 248056,
|
| 6 |
"model_type": "qwen3_5_moe",
|
| 7 |
+
"quantization_config": {
|
| 8 |
+
"config_groups": {
|
| 9 |
+
"group_0": {
|
| 10 |
+
"format": "float-quantized",
|
| 11 |
+
"input_activations": {
|
| 12 |
+
"actorder": null,
|
| 13 |
+
"block_structure": null,
|
| 14 |
+
"dynamic": true,
|
| 15 |
+
"group_size": null,
|
| 16 |
+
"num_bits": 8,
|
| 17 |
+
"observer": null,
|
| 18 |
+
"observer_kwargs": {},
|
| 19 |
+
"scale_dtype": null,
|
| 20 |
+
"strategy": "token",
|
| 21 |
+
"symmetric": true,
|
| 22 |
+
"type": "float",
|
| 23 |
+
"zp_dtype": null
|
| 24 |
+
},
|
| 25 |
+
"output_activations": null,
|
| 26 |
+
"targets": [
|
| 27 |
+
"Linear"
|
| 28 |
+
],
|
| 29 |
+
"weights": {
|
| 30 |
+
"actorder": null,
|
| 31 |
+
"block_structure": null,
|
| 32 |
+
"dynamic": false,
|
| 33 |
+
"group_size": null,
|
| 34 |
+
"num_bits": 8,
|
| 35 |
+
"observer": "memoryless_minmax",
|
| 36 |
+
"observer_kwargs": {},
|
| 37 |
+
"scale_dtype": null,
|
| 38 |
+
"strategy": "channel",
|
| 39 |
+
"symmetric": true,
|
| 40 |
+
"type": "float",
|
| 41 |
+
"zp_dtype": null
|
| 42 |
+
}
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"format": "float-quantized",
|
| 46 |
+
"global_compression_ratio": null,
|
| 47 |
+
"ignore": [
|
| 48 |
+
"model.visual.blocks.0.attn.qkv",
|
| 49 |
+
"model.visual.blocks.0.attn.proj",
|
| 50 |
+
"model.visual.blocks.0.mlp.linear_fc1",
|
| 51 |
+
"model.visual.blocks.0.mlp.linear_fc2",
|
| 52 |
+
"model.visual.blocks.1.attn.qkv",
|
| 53 |
+
"model.visual.blocks.1.attn.proj",
|
| 54 |
+
"model.visual.blocks.1.mlp.linear_fc1",
|
| 55 |
+
"model.visual.blocks.1.mlp.linear_fc2",
|
| 56 |
+
"model.visual.blocks.2.attn.qkv",
|
| 57 |
+
"model.visual.blocks.2.attn.proj",
|
| 58 |
+
"model.visual.blocks.2.mlp.linear_fc1",
|
| 59 |
+
"model.visual.blocks.2.mlp.linear_fc2",
|
| 60 |
+
"model.visual.blocks.3.attn.qkv",
|
| 61 |
+
"model.visual.blocks.3.attn.proj",
|
| 62 |
+
"model.visual.blocks.3.mlp.linear_fc1",
|
| 63 |
+
"model.visual.blocks.3.mlp.linear_fc2",
|
| 64 |
+
"model.visual.blocks.4.attn.qkv",
|
| 65 |
+
"model.visual.blocks.4.attn.proj",
|
| 66 |
+
"model.visual.blocks.4.mlp.linear_fc1",
|
| 67 |
+
"model.visual.blocks.4.mlp.linear_fc2",
|
| 68 |
+
"model.visual.blocks.5.attn.qkv",
|
| 69 |
+
"model.visual.blocks.5.attn.proj",
|
| 70 |
+
"model.visual.blocks.5.mlp.linear_fc1",
|
| 71 |
+
"model.visual.blocks.5.mlp.linear_fc2",
|
| 72 |
+
"model.visual.blocks.6.attn.qkv",
|
| 73 |
+
"model.visual.blocks.6.attn.proj",
|
| 74 |
+
"model.visual.blocks.6.mlp.linear_fc1",
|
| 75 |
+
"model.visual.blocks.6.mlp.linear_fc2",
|
| 76 |
+
"model.visual.blocks.7.attn.qkv",
|
| 77 |
+
"model.visual.blocks.7.attn.proj",
|
| 78 |
+
"model.visual.blocks.7.mlp.linear_fc1",
|
| 79 |
+
"model.visual.blocks.7.mlp.linear_fc2",
|
| 80 |
+
"model.visual.blocks.8.attn.qkv",
|
| 81 |
+
"model.visual.blocks.8.attn.proj",
|
| 82 |
+
"model.visual.blocks.8.mlp.linear_fc1",
|
| 83 |
+
"model.visual.blocks.8.mlp.linear_fc2",
|
| 84 |
+
"model.visual.blocks.9.attn.qkv",
|
| 85 |
+
"model.visual.blocks.9.attn.proj",
|
| 86 |
+
"model.visual.blocks.9.mlp.linear_fc1",
|
| 87 |
+
"model.visual.blocks.9.mlp.linear_fc2",
|
| 88 |
+
"model.visual.blocks.10.attn.qkv",
|
| 89 |
+
"model.visual.blocks.10.attn.proj",
|
| 90 |
+
"model.visual.blocks.10.mlp.linear_fc1",
|
| 91 |
+
"model.visual.blocks.10.mlp.linear_fc2",
|
| 92 |
+
"model.visual.blocks.11.attn.qkv",
|
| 93 |
+
"model.visual.blocks.11.attn.proj",
|
| 94 |
+
"model.visual.blocks.11.mlp.linear_fc1",
|
| 95 |
+
"model.visual.blocks.11.mlp.linear_fc2",
|
| 96 |
+
"model.visual.blocks.12.attn.qkv",
|
| 97 |
+
"model.visual.blocks.12.attn.proj",
|
| 98 |
+
"model.visual.blocks.12.mlp.linear_fc1",
|
| 99 |
+
"model.visual.blocks.12.mlp.linear_fc2",
|
| 100 |
+
"model.visual.blocks.13.attn.qkv",
|
| 101 |
+
"model.visual.blocks.13.attn.proj",
|
| 102 |
+
"model.visual.blocks.13.mlp.linear_fc1",
|
| 103 |
+
"model.visual.blocks.13.mlp.linear_fc2",
|
| 104 |
+
"model.visual.blocks.14.attn.qkv",
|
| 105 |
+
"model.visual.blocks.14.attn.proj",
|
| 106 |
+
"model.visual.blocks.14.mlp.linear_fc1",
|
| 107 |
+
"model.visual.blocks.14.mlp.linear_fc2",
|
| 108 |
+
"model.visual.blocks.15.attn.qkv",
|
| 109 |
+
"model.visual.blocks.15.attn.proj",
|
| 110 |
+
"model.visual.blocks.15.mlp.linear_fc1",
|
| 111 |
+
"model.visual.blocks.15.mlp.linear_fc2",
|
| 112 |
+
"model.visual.blocks.16.attn.qkv",
|
| 113 |
+
"model.visual.blocks.16.attn.proj",
|
| 114 |
+
"model.visual.blocks.16.mlp.linear_fc1",
|
| 115 |
+
"model.visual.blocks.16.mlp.linear_fc2",
|
| 116 |
+
"model.visual.blocks.17.attn.qkv",
|
| 117 |
+
"model.visual.blocks.17.attn.proj",
|
| 118 |
+
"model.visual.blocks.17.mlp.linear_fc1",
|
| 119 |
+
"model.visual.blocks.17.mlp.linear_fc2",
|
| 120 |
+
"model.visual.blocks.18.attn.qkv",
|
| 121 |
+
"model.visual.blocks.18.attn.proj",
|
| 122 |
+
"model.visual.blocks.18.mlp.linear_fc1",
|
| 123 |
+
"model.visual.blocks.18.mlp.linear_fc2",
|
| 124 |
+
"model.visual.blocks.19.attn.qkv",
|
| 125 |
+
"model.visual.blocks.19.attn.proj",
|
| 126 |
+
"model.visual.blocks.19.mlp.linear_fc1",
|
| 127 |
+
"model.visual.blocks.19.mlp.linear_fc2",
|
| 128 |
+
"model.visual.blocks.20.attn.qkv",
|
| 129 |
+
"model.visual.blocks.20.attn.proj",
|
| 130 |
+
"model.visual.blocks.20.mlp.linear_fc1",
|
| 131 |
+
"model.visual.blocks.20.mlp.linear_fc2",
|
| 132 |
+
"model.visual.blocks.21.attn.qkv",
|
| 133 |
+
"model.visual.blocks.21.attn.proj",
|
| 134 |
+
"model.visual.blocks.21.mlp.linear_fc1",
|
| 135 |
+
"model.visual.blocks.21.mlp.linear_fc2",
|
| 136 |
+
"model.visual.blocks.22.attn.qkv",
|
| 137 |
+
"model.visual.blocks.22.attn.proj",
|
| 138 |
+
"model.visual.blocks.22.mlp.linear_fc1",
|
| 139 |
+
"model.visual.blocks.22.mlp.linear_fc2",
|
| 140 |
+
"model.visual.blocks.23.attn.qkv",
|
| 141 |
+
"model.visual.blocks.23.attn.proj",
|
| 142 |
+
"model.visual.blocks.23.mlp.linear_fc1",
|
| 143 |
+
"model.visual.blocks.23.mlp.linear_fc2",
|
| 144 |
+
"model.visual.blocks.24.attn.qkv",
|
| 145 |
+
"model.visual.blocks.24.attn.proj",
|
| 146 |
+
"model.visual.blocks.24.mlp.linear_fc1",
|
| 147 |
+
"model.visual.blocks.24.mlp.linear_fc2",
|
| 148 |
+
"model.visual.blocks.25.attn.qkv",
|
| 149 |
+
"model.visual.blocks.25.attn.proj",
|
| 150 |
+
"model.visual.blocks.25.mlp.linear_fc1",
|
| 151 |
+
"model.visual.blocks.25.mlp.linear_fc2",
|
| 152 |
+
"model.visual.blocks.26.attn.qkv",
|
| 153 |
+
"model.visual.blocks.26.attn.proj",
|
| 154 |
+
"model.visual.blocks.26.mlp.linear_fc1",
|
| 155 |
+
"model.visual.blocks.26.mlp.linear_fc2",
|
| 156 |
+
"model.visual.merger.linear_fc1",
|
| 157 |
+
"model.visual.merger.linear_fc2",
|
| 158 |
+
"model.language_model.layers.0.linear_attn.out_proj",
|
| 159 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv",
|
| 160 |
+
"model.language_model.layers.0.linear_attn.in_proj_z",
|
| 161 |
+
"model.language_model.layers.0.linear_attn.in_proj_b",
|
| 162 |
+
"model.language_model.layers.0.linear_attn.in_proj_a",
|
| 163 |
+
"model.language_model.layers.0.mlp.shared_expert_gate",
|
| 164 |
+
"model.language_model.layers.1.linear_attn.out_proj",
|
| 165 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv",
|
| 166 |
+
"model.language_model.layers.1.linear_attn.in_proj_z",
|
| 167 |
+
"model.language_model.layers.1.linear_attn.in_proj_b",
|
| 168 |
+
"model.language_model.layers.1.linear_attn.in_proj_a",
|
| 169 |
+
"model.language_model.layers.1.mlp.shared_expert_gate",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.out_proj",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_z",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_b",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_a",
|
| 175 |
+
"model.language_model.layers.2.mlp.shared_expert_gate",
|
| 176 |
+
"model.language_model.layers.3.mlp.shared_expert_gate",
|
| 177 |
+
"model.language_model.layers.4.linear_attn.out_proj",
|
| 178 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv",
|
| 179 |
+
"model.language_model.layers.4.linear_attn.in_proj_z",
|
| 180 |
+
"model.language_model.layers.4.linear_attn.in_proj_b",
|
| 181 |
+
"model.language_model.layers.4.linear_attn.in_proj_a",
|
| 182 |
+
"model.language_model.layers.4.mlp.shared_expert_gate",
|
| 183 |
+
"model.language_model.layers.5.linear_attn.out_proj",
|
| 184 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv",
|
| 185 |
+
"model.language_model.layers.5.linear_attn.in_proj_z",
|
| 186 |
+
"model.language_model.layers.5.linear_attn.in_proj_b",
|
| 187 |
+
"model.language_model.layers.5.linear_attn.in_proj_a",
|
| 188 |
+
"model.language_model.layers.5.mlp.shared_expert_gate",
|
| 189 |
+
"model.language_model.layers.6.linear_attn.out_proj",
|
| 190 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv",
|
| 191 |
+
"model.language_model.layers.6.linear_attn.in_proj_z",
|
| 192 |
+
"model.language_model.layers.6.linear_attn.in_proj_b",
|
| 193 |
+
"model.language_model.layers.6.linear_attn.in_proj_a",
|
| 194 |
+
"model.language_model.layers.6.mlp.shared_expert_gate",
|
| 195 |
+
"model.language_model.layers.7.mlp.shared_expert_gate",
|
| 196 |
+
"model.language_model.layers.8.linear_attn.out_proj",
|
| 197 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv",
|
| 198 |
+
"model.language_model.layers.8.linear_attn.in_proj_z",
|
| 199 |
+
"model.language_model.layers.8.linear_attn.in_proj_b",
|
| 200 |
+
"model.language_model.layers.8.linear_attn.in_proj_a",
|
| 201 |
+
"model.language_model.layers.8.mlp.shared_expert_gate",
|
| 202 |
+
"model.language_model.layers.9.linear_attn.out_proj",
|
| 203 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv",
|
| 204 |
+
"model.language_model.layers.9.linear_attn.in_proj_z",
|
| 205 |
+
"model.language_model.layers.9.linear_attn.in_proj_b",
|
| 206 |
+
"model.language_model.layers.9.linear_attn.in_proj_a",
|
| 207 |
+
"model.language_model.layers.9.mlp.shared_expert_gate",
|
| 208 |
+
"model.language_model.layers.10.linear_attn.out_proj",
|
| 209 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv",
|
| 210 |
+
"model.language_model.layers.10.linear_attn.in_proj_z",
|
| 211 |
+
"model.language_model.layers.10.linear_attn.in_proj_b",
|
| 212 |
+
"model.language_model.layers.10.linear_attn.in_proj_a",
|
| 213 |
+
"model.language_model.layers.10.mlp.shared_expert_gate",
|
| 214 |
+
"model.language_model.layers.11.mlp.shared_expert_gate",
|
| 215 |
+
"model.language_model.layers.12.linear_attn.out_proj",
|
| 216 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv",
|
| 217 |
+
"model.language_model.layers.12.linear_attn.in_proj_z",
|
| 218 |
+
"model.language_model.layers.12.linear_attn.in_proj_b",
|
| 219 |
+
"model.language_model.layers.12.linear_attn.in_proj_a",
|
| 220 |
+
"model.language_model.layers.12.mlp.shared_expert_gate",
|
| 221 |
+
"model.language_model.layers.13.linear_attn.out_proj",
|
| 222 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv",
|
| 223 |
+
"model.language_model.layers.13.linear_attn.in_proj_z",
|
| 224 |
+
"model.language_model.layers.13.linear_attn.in_proj_b",
|
| 225 |
+
"model.language_model.layers.13.linear_attn.in_proj_a",
|
| 226 |
+
"model.language_model.layers.13.mlp.shared_expert_gate",
|
| 227 |
+
"model.language_model.layers.14.linear_attn.out_proj",
|
| 228 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv",
|
| 229 |
+
"model.language_model.layers.14.linear_attn.in_proj_z",
|
| 230 |
+
"model.language_model.layers.14.linear_attn.in_proj_b",
|
| 231 |
+
"model.language_model.layers.14.linear_attn.in_proj_a",
|
| 232 |
+
"model.language_model.layers.14.mlp.shared_expert_gate",
|
| 233 |
+
"model.language_model.layers.15.mlp.shared_expert_gate",
|
| 234 |
+
"model.language_model.layers.16.linear_attn.out_proj",
|
| 235 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv",
|
| 236 |
+
"model.language_model.layers.16.linear_attn.in_proj_z",
|
| 237 |
+
"model.language_model.layers.16.linear_attn.in_proj_b",
|
| 238 |
+
"model.language_model.layers.16.linear_attn.in_proj_a",
|
| 239 |
+
"model.language_model.layers.16.mlp.shared_expert_gate",
|
| 240 |
+
"model.language_model.layers.17.linear_attn.out_proj",
|
| 241 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv",
|
| 242 |
+
"model.language_model.layers.17.linear_attn.in_proj_z",
|
| 243 |
+
"model.language_model.layers.17.linear_attn.in_proj_b",
|
| 244 |
+
"model.language_model.layers.17.linear_attn.in_proj_a",
|
| 245 |
+
"model.language_model.layers.17.mlp.shared_expert_gate",
|
| 246 |
+
"model.language_model.layers.18.linear_attn.out_proj",
|
| 247 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv",
|
| 248 |
+
"model.language_model.layers.18.linear_attn.in_proj_z",
|
| 249 |
+
"model.language_model.layers.18.linear_attn.in_proj_b",
|
| 250 |
+
"model.language_model.layers.18.linear_attn.in_proj_a",
|
| 251 |
+
"model.language_model.layers.18.mlp.shared_expert_gate",
|
| 252 |
+
"model.language_model.layers.19.mlp.shared_expert_gate",
|
| 253 |
+
"model.language_model.layers.20.linear_attn.out_proj",
|
| 254 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv",
|
| 255 |
+
"model.language_model.layers.20.linear_attn.in_proj_z",
|
| 256 |
+
"model.language_model.layers.20.linear_attn.in_proj_b",
|
| 257 |
+
"model.language_model.layers.20.linear_attn.in_proj_a",
|
| 258 |
+
"model.language_model.layers.20.mlp.shared_expert_gate",
|
| 259 |
+
"model.language_model.layers.21.linear_attn.out_proj",
|
| 260 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv",
|
| 261 |
+
"model.language_model.layers.21.linear_attn.in_proj_z",
|
| 262 |
+
"model.language_model.layers.21.linear_attn.in_proj_b",
|
| 263 |
+
"model.language_model.layers.21.linear_attn.in_proj_a",
|
| 264 |
+
"model.language_model.layers.21.mlp.shared_expert_gate",
|
| 265 |
+
"model.language_model.layers.22.linear_attn.out_proj",
|
| 266 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv",
|
| 267 |
+
"model.language_model.layers.22.linear_attn.in_proj_z",
|
| 268 |
+
"model.language_model.layers.22.linear_attn.in_proj_b",
|
| 269 |
+
"model.language_model.layers.22.linear_attn.in_proj_a",
|
| 270 |
+
"model.language_model.layers.22.mlp.shared_expert_gate",
|
| 271 |
+
"model.language_model.layers.23.mlp.shared_expert_gate",
|
| 272 |
+
"model.language_model.layers.24.linear_attn.out_proj",
|
| 273 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv",
|
| 274 |
+
"model.language_model.layers.24.linear_attn.in_proj_z",
|
| 275 |
+
"model.language_model.layers.24.linear_attn.in_proj_b",
|
| 276 |
+
"model.language_model.layers.24.linear_attn.in_proj_a",
|
| 277 |
+
"model.language_model.layers.24.mlp.shared_expert_gate",
|
| 278 |
+
"model.language_model.layers.25.linear_attn.out_proj",
|
| 279 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv",
|
| 280 |
+
"model.language_model.layers.25.linear_attn.in_proj_z",
|
| 281 |
+
"model.language_model.layers.25.linear_attn.in_proj_b",
|
| 282 |
+
"model.language_model.layers.25.linear_attn.in_proj_a",
|
| 283 |
+
"model.language_model.layers.25.mlp.shared_expert_gate",
|
| 284 |
+
"model.language_model.layers.26.linear_attn.out_proj",
|
| 285 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv",
|
| 286 |
+
"model.language_model.layers.26.linear_attn.in_proj_z",
|
| 287 |
+
"model.language_model.layers.26.linear_attn.in_proj_b",
|
| 288 |
+
"model.language_model.layers.26.linear_attn.in_proj_a",
|
| 289 |
+
"model.language_model.layers.26.mlp.shared_expert_gate",
|
| 290 |
+
"model.language_model.layers.27.mlp.shared_expert_gate",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.out_proj",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_z",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_b",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_a",
|
| 296 |
+
"model.language_model.layers.28.mlp.shared_expert_gate",
|
| 297 |
+
"model.language_model.layers.29.linear_attn.out_proj",
|
| 298 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv",
|
| 299 |
+
"model.language_model.layers.29.linear_attn.in_proj_z",
|
| 300 |
+
"model.language_model.layers.29.linear_attn.in_proj_b",
|
| 301 |
+
"model.language_model.layers.29.linear_attn.in_proj_a",
|
| 302 |
+
"model.language_model.layers.29.mlp.shared_expert_gate",
|
| 303 |
+
"model.language_model.layers.30.linear_attn.out_proj",
|
| 304 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv",
|
| 305 |
+
"model.language_model.layers.30.linear_attn.in_proj_z",
|
| 306 |
+
"model.language_model.layers.30.linear_attn.in_proj_b",
|
| 307 |
+
"model.language_model.layers.30.linear_attn.in_proj_a",
|
| 308 |
+
"model.language_model.layers.30.mlp.shared_expert_gate",
|
| 309 |
+
"model.language_model.layers.31.mlp.shared_expert_gate",
|
| 310 |
+
"model.language_model.layers.32.linear_attn.out_proj",
|
| 311 |
+
"model.language_model.layers.32.linear_attn.in_proj_qkv",
|
| 312 |
+
"model.language_model.layers.32.linear_attn.in_proj_z",
|
| 313 |
+
"model.language_model.layers.32.linear_attn.in_proj_b",
|
| 314 |
+
"model.language_model.layers.32.linear_attn.in_proj_a",
|
| 315 |
+
"model.language_model.layers.32.mlp.shared_expert_gate",
|
| 316 |
+
"model.language_model.layers.33.linear_attn.out_proj",
|
| 317 |
+
"model.language_model.layers.33.linear_attn.in_proj_qkv",
|
| 318 |
+
"model.language_model.layers.33.linear_attn.in_proj_z",
|
| 319 |
+
"model.language_model.layers.33.linear_attn.in_proj_b",
|
| 320 |
+
"model.language_model.layers.33.linear_attn.in_proj_a",
|
| 321 |
+
"model.language_model.layers.33.mlp.shared_expert_gate",
|
| 322 |
+
"model.language_model.layers.34.linear_attn.out_proj",
|
| 323 |
+
"model.language_model.layers.34.linear_attn.in_proj_qkv",
|
| 324 |
+
"model.language_model.layers.34.linear_attn.in_proj_z",
|
| 325 |
+
"model.language_model.layers.34.linear_attn.in_proj_b",
|
| 326 |
+
"model.language_model.layers.34.linear_attn.in_proj_a",
|
| 327 |
+
"model.language_model.layers.34.mlp.shared_expert_gate",
|
| 328 |
+
"model.language_model.layers.35.mlp.shared_expert_gate",
|
| 329 |
+
"model.language_model.layers.36.linear_attn.out_proj",
|
| 330 |
+
"model.language_model.layers.36.linear_attn.in_proj_qkv",
|
| 331 |
+
"model.language_model.layers.36.linear_attn.in_proj_z",
|
| 332 |
+
"model.language_model.layers.36.linear_attn.in_proj_b",
|
| 333 |
+
"model.language_model.layers.36.linear_attn.in_proj_a",
|
| 334 |
+
"model.language_model.layers.36.mlp.shared_expert_gate",
|
| 335 |
+
"model.language_model.layers.37.linear_attn.out_proj",
|
| 336 |
+
"model.language_model.layers.37.linear_attn.in_proj_qkv",
|
| 337 |
+
"model.language_model.layers.37.linear_attn.in_proj_z",
|
| 338 |
+
"model.language_model.layers.37.linear_attn.in_proj_b",
|
| 339 |
+
"model.language_model.layers.37.linear_attn.in_proj_a",
|
| 340 |
+
"model.language_model.layers.37.mlp.shared_expert_gate",
|
| 341 |
+
"model.language_model.layers.38.linear_attn.out_proj",
|
| 342 |
+
"model.language_model.layers.38.linear_attn.in_proj_qkv",
|
| 343 |
+
"model.language_model.layers.38.linear_attn.in_proj_z",
|
| 344 |
+
"model.language_model.layers.38.linear_attn.in_proj_b",
|
| 345 |
+
"model.language_model.layers.38.linear_attn.in_proj_a",
|
| 346 |
+
"model.language_model.layers.38.mlp.shared_expert_gate",
|
| 347 |
+
"model.language_model.layers.39.mlp.shared_expert_gate",
|
| 348 |
+
"model.language_model.layers.40.linear_attn.out_proj",
|
| 349 |
+
"model.language_model.layers.40.linear_attn.in_proj_qkv",
|
| 350 |
+
"model.language_model.layers.40.linear_attn.in_proj_z",
|
| 351 |
+
"model.language_model.layers.40.linear_attn.in_proj_b",
|
| 352 |
+
"model.language_model.layers.40.linear_attn.in_proj_a",
|
| 353 |
+
"model.language_model.layers.40.mlp.shared_expert_gate",
|
| 354 |
+
"model.language_model.layers.41.linear_attn.out_proj",
|
| 355 |
+
"model.language_model.layers.41.linear_attn.in_proj_qkv",
|
| 356 |
+
"model.language_model.layers.41.linear_attn.in_proj_z",
|
| 357 |
+
"model.language_model.layers.41.linear_attn.in_proj_b",
|
| 358 |
+
"model.language_model.layers.41.linear_attn.in_proj_a",
|
| 359 |
+
"model.language_model.layers.41.mlp.shared_expert_gate",
|
| 360 |
+
"model.language_model.layers.42.linear_attn.out_proj",
|
| 361 |
+
"model.language_model.layers.42.linear_attn.in_proj_qkv",
|
| 362 |
+
"model.language_model.layers.42.linear_attn.in_proj_z",
|
| 363 |
+
"model.language_model.layers.42.linear_attn.in_proj_b",
|
| 364 |
+
"model.language_model.layers.42.linear_attn.in_proj_a",
|
| 365 |
+
"model.language_model.layers.42.mlp.shared_expert_gate",
|
| 366 |
+
"model.language_model.layers.43.mlp.shared_expert_gate",
|
| 367 |
+
"model.language_model.layers.44.linear_attn.out_proj",
|
| 368 |
+
"model.language_model.layers.44.linear_attn.in_proj_qkv",
|
| 369 |
+
"model.language_model.layers.44.linear_attn.in_proj_z",
|
| 370 |
+
"model.language_model.layers.44.linear_attn.in_proj_b",
|
| 371 |
+
"model.language_model.layers.44.linear_attn.in_proj_a",
|
| 372 |
+
"model.language_model.layers.44.mlp.shared_expert_gate",
|
| 373 |
+
"model.language_model.layers.45.linear_attn.out_proj",
|
| 374 |
+
"model.language_model.layers.45.linear_attn.in_proj_qkv",
|
| 375 |
+
"model.language_model.layers.45.linear_attn.in_proj_z",
|
| 376 |
+
"model.language_model.layers.45.linear_attn.in_proj_b",
|
| 377 |
+
"model.language_model.layers.45.linear_attn.in_proj_a",
|
| 378 |
+
"model.language_model.layers.45.mlp.shared_expert_gate",
|
| 379 |
+
"model.language_model.layers.46.linear_attn.out_proj",
|
| 380 |
+
"model.language_model.layers.46.linear_attn.in_proj_qkv",
|
| 381 |
+
"model.language_model.layers.46.linear_attn.in_proj_z",
|
| 382 |
+
"model.language_model.layers.46.linear_attn.in_proj_b",
|
| 383 |
+
"model.language_model.layers.46.linear_attn.in_proj_a",
|
| 384 |
+
"model.language_model.layers.46.mlp.shared_expert_gate",
|
| 385 |
+
"model.language_model.layers.47.mlp.shared_expert_gate",
|
| 386 |
+
"model.language_model.layers.48.linear_attn.out_proj",
|
| 387 |
+
"model.language_model.layers.48.linear_attn.in_proj_qkv",
|
| 388 |
+
"model.language_model.layers.48.linear_attn.in_proj_z",
|
| 389 |
+
"model.language_model.layers.48.linear_attn.in_proj_b",
|
| 390 |
+
"model.language_model.layers.48.linear_attn.in_proj_a",
|
| 391 |
+
"model.language_model.layers.48.mlp.shared_expert_gate",
|
| 392 |
+
"model.language_model.layers.49.linear_attn.out_proj",
|
| 393 |
+
"model.language_model.layers.49.linear_attn.in_proj_qkv",
|
| 394 |
+
"model.language_model.layers.49.linear_attn.in_proj_z",
|
| 395 |
+
"model.language_model.layers.49.linear_attn.in_proj_b",
|
| 396 |
+
"model.language_model.layers.49.linear_attn.in_proj_a",
|
| 397 |
+
"model.language_model.layers.49.mlp.shared_expert_gate",
|
| 398 |
+
"model.language_model.layers.50.linear_attn.out_proj",
|
| 399 |
+
"model.language_model.layers.50.linear_attn.in_proj_qkv",
|
| 400 |
+
"model.language_model.layers.50.linear_attn.in_proj_z",
|
| 401 |
+
"model.language_model.layers.50.linear_attn.in_proj_b",
|
| 402 |
+
"model.language_model.layers.50.linear_attn.in_proj_a",
|
| 403 |
+
"model.language_model.layers.50.mlp.shared_expert_gate",
|
| 404 |
+
"model.language_model.layers.51.mlp.shared_expert_gate",
|
| 405 |
+
"model.language_model.layers.52.linear_attn.out_proj",
|
| 406 |
+
"model.language_model.layers.52.linear_attn.in_proj_qkv",
|
| 407 |
+
"model.language_model.layers.52.linear_attn.in_proj_z",
|
| 408 |
+
"model.language_model.layers.52.linear_attn.in_proj_b",
|
| 409 |
+
"model.language_model.layers.52.linear_attn.in_proj_a",
|
| 410 |
+
"model.language_model.layers.52.mlp.shared_expert_gate",
|
| 411 |
+
"model.language_model.layers.53.linear_attn.out_proj",
|
| 412 |
+
"model.language_model.layers.53.linear_attn.in_proj_qkv",
|
| 413 |
+
"model.language_model.layers.53.linear_attn.in_proj_z",
|
| 414 |
+
"model.language_model.layers.53.linear_attn.in_proj_b",
|
| 415 |
+
"model.language_model.layers.53.linear_attn.in_proj_a",
|
| 416 |
+
"model.language_model.layers.53.mlp.shared_expert_gate",
|
| 417 |
+
"model.language_model.layers.54.linear_attn.out_proj",
|
| 418 |
+
"model.language_model.layers.54.linear_attn.in_proj_qkv",
|
| 419 |
+
"model.language_model.layers.54.linear_attn.in_proj_z",
|
| 420 |
+
"model.language_model.layers.54.linear_attn.in_proj_b",
|
| 421 |
+
"model.language_model.layers.54.linear_attn.in_proj_a",
|
| 422 |
+
"model.language_model.layers.54.mlp.shared_expert_gate",
|
| 423 |
+
"model.language_model.layers.55.mlp.shared_expert_gate",
|
| 424 |
+
"model.language_model.layers.56.linear_attn.out_proj",
|
| 425 |
+
"model.language_model.layers.56.linear_attn.in_proj_qkv",
|
| 426 |
+
"model.language_model.layers.56.linear_attn.in_proj_z",
|
| 427 |
+
"model.language_model.layers.56.linear_attn.in_proj_b",
|
| 428 |
+
"model.language_model.layers.56.linear_attn.in_proj_a",
|
| 429 |
+
"model.language_model.layers.56.mlp.shared_expert_gate",
|
| 430 |
+
"model.language_model.layers.57.linear_attn.out_proj",
|
| 431 |
+
"model.language_model.layers.57.linear_attn.in_proj_qkv",
|
| 432 |
+
"model.language_model.layers.57.linear_attn.in_proj_z",
|
| 433 |
+
"model.language_model.layers.57.linear_attn.in_proj_b",
|
| 434 |
+
"model.language_model.layers.57.linear_attn.in_proj_a",
|
| 435 |
+
"model.language_model.layers.57.mlp.shared_expert_gate",
|
| 436 |
+
"model.language_model.layers.58.linear_attn.out_proj",
|
| 437 |
+
"model.language_model.layers.58.linear_attn.in_proj_qkv",
|
| 438 |
+
"model.language_model.layers.58.linear_attn.in_proj_z",
|
| 439 |
+
"model.language_model.layers.58.linear_attn.in_proj_b",
|
| 440 |
+
"model.language_model.layers.58.linear_attn.in_proj_a",
|
| 441 |
+
"model.language_model.layers.58.mlp.shared_expert_gate",
|
| 442 |
+
"model.language_model.layers.59.mlp.shared_expert_gate",
|
| 443 |
+
"model.language_model.layers.0.mlp.gate",
|
| 444 |
+
"model.language_model.layers.1.mlp.gate",
|
| 445 |
+
"model.language_model.layers.2.mlp.gate",
|
| 446 |
+
"model.language_model.layers.3.mlp.gate",
|
| 447 |
+
"model.language_model.layers.4.mlp.gate",
|
| 448 |
+
"model.language_model.layers.5.mlp.gate",
|
| 449 |
+
"model.language_model.layers.6.mlp.gate",
|
| 450 |
+
"model.language_model.layers.7.mlp.gate",
|
| 451 |
+
"model.language_model.layers.8.mlp.gate",
|
| 452 |
+
"model.language_model.layers.9.mlp.gate",
|
| 453 |
+
"model.language_model.layers.10.mlp.gate",
|
| 454 |
+
"model.language_model.layers.11.mlp.gate",
|
| 455 |
+
"model.language_model.layers.12.mlp.gate",
|
| 456 |
+
"model.language_model.layers.13.mlp.gate",
|
| 457 |
+
"model.language_model.layers.14.mlp.gate",
|
| 458 |
+
"model.language_model.layers.15.mlp.gate",
|
| 459 |
+
"model.language_model.layers.16.mlp.gate",
|
| 460 |
+
"model.language_model.layers.17.mlp.gate",
|
| 461 |
+
"model.language_model.layers.18.mlp.gate",
|
| 462 |
+
"model.language_model.layers.19.mlp.gate",
|
| 463 |
+
"model.language_model.layers.20.mlp.gate",
|
| 464 |
+
"model.language_model.layers.21.mlp.gate",
|
| 465 |
+
"model.language_model.layers.22.mlp.gate",
|
| 466 |
+
"model.language_model.layers.23.mlp.gate",
|
| 467 |
+
"model.language_model.layers.24.mlp.gate",
|
| 468 |
+
"model.language_model.layers.25.mlp.gate",
|
| 469 |
+
"model.language_model.layers.26.mlp.gate",
|
| 470 |
+
"model.language_model.layers.27.mlp.gate",
|
| 471 |
+
"model.language_model.layers.28.mlp.gate",
|
| 472 |
+
"model.language_model.layers.29.mlp.gate",
|
| 473 |
+
"model.language_model.layers.30.mlp.gate",
|
| 474 |
+
"model.language_model.layers.31.mlp.gate",
|
| 475 |
+
"model.language_model.layers.32.mlp.gate",
|
| 476 |
+
"model.language_model.layers.33.mlp.gate",
|
| 477 |
+
"model.language_model.layers.34.mlp.gate",
|
| 478 |
+
"model.language_model.layers.35.mlp.gate",
|
| 479 |
+
"model.language_model.layers.36.mlp.gate",
|
| 480 |
+
"model.language_model.layers.37.mlp.gate",
|
| 481 |
+
"model.language_model.layers.38.mlp.gate",
|
| 482 |
+
"model.language_model.layers.39.mlp.gate",
|
| 483 |
+
"model.language_model.layers.40.mlp.gate",
|
| 484 |
+
"model.language_model.layers.41.mlp.gate",
|
| 485 |
+
"model.language_model.layers.42.mlp.gate",
|
| 486 |
+
"model.language_model.layers.43.mlp.gate",
|
| 487 |
+
"model.language_model.layers.44.mlp.gate",
|
| 488 |
+
"model.language_model.layers.45.mlp.gate",
|
| 489 |
+
"model.language_model.layers.46.mlp.gate",
|
| 490 |
+
"model.language_model.layers.47.mlp.gate",
|
| 491 |
+
"model.language_model.layers.48.mlp.gate",
|
| 492 |
+
"model.language_model.layers.49.mlp.gate",
|
| 493 |
+
"model.language_model.layers.50.mlp.gate",
|
| 494 |
+
"model.language_model.layers.51.mlp.gate",
|
| 495 |
+
"model.language_model.layers.52.mlp.gate",
|
| 496 |
+
"model.language_model.layers.53.mlp.gate",
|
| 497 |
+
"model.language_model.layers.54.mlp.gate",
|
| 498 |
+
"model.language_model.layers.55.mlp.gate",
|
| 499 |
+
"model.language_model.layers.56.mlp.gate",
|
| 500 |
+
"model.language_model.layers.57.mlp.gate",
|
| 501 |
+
"model.language_model.layers.58.mlp.gate",
|
| 502 |
+
"model.language_model.layers.59.mlp.gate",
|
| 503 |
+
"lm_head"
|
| 504 |
+
],
|
| 505 |
+
"kv_cache_scheme": null,
|
| 506 |
+
"quant_method": "compressed-tensors",
|
| 507 |
+
"quantization_status": "compressed",
|
| 508 |
+
"sparsity_config": {},
|
| 509 |
+
"transform_config": {},
|
| 510 |
+
"version": "0.13.1.a20260212"
|
| 511 |
+
},
|
| 512 |
"text_config": {
|
| 513 |
"attention_bias": false,
|
| 514 |
"attention_dropout": 0.0,
|
|
|
|
| 636 |
"temporal_patch_size": 2
|
| 637 |
},
|
| 638 |
"vision_end_token_id": 248054,
|
| 639 |
+
"vision_start_token_id": 248053
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 640 |
}
|
model-00001-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:792cf5a7d4e617f4d9d81d78867eca7f5837ee21fb8629ec1490165d7074e014
|
| 3 |
+
size 49999213680
|
model-00002-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20c65bc753d98a64c141fea0cbd0c82da2bc0771da96b840899e6ab339ca1d73
|
| 3 |
+
size 49999230608
|
model-00003-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ee682504f43b6478edcf5cd7f26e72799612b752bef064e2df03f71b6cb4e68
|
| 3 |
+
size 50002504832
|
model-00004-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e2297f2c996ecaa527037476f6569e99a0886a52a34fdabef8a91253ad93af8
|
| 3 |
+
size 49999233960
|
model-00005-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:118748a384ecb33fc1d67d1d8549d547d9c2d29eda9174ca0d28457807e9798d
|
| 3 |
+
size 50002509184
|
model-00006-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d82cab0327c3ef4b9b1725c4a7ddd3b8ad5fe541f1ff897aec5ed609f2a164f
|
| 3 |
+
size 49999240104
|
model-00007-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d070b1a7c0778ae4e99994cf17ce710320cd58fd2cafd53eb517c7cac29b9d0
|
| 3 |
+
size 50002571416
|
model-00008-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa3265eedf9b295b39be2f64ba9f13f19637377c3dc8161e62821dd2a7996ae
|
| 3 |
+
size 49999240304
|
model-00009-of-00009.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b6ab3479dae252e850a6516930ed0ebca2b6c451cdefc7227c2372a1749c5f3
|
| 3 |
+
size 5131688016
|
model.safetensors.index.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82f385c18828151bb1e15a32b80870183f4711283a52d03fc86353bebe27ee9e
|
| 3 |
+
size 20228802
|
old_config.json
ADDED
|
@@ -0,0 +1,647 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5MoeForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"image_token_id": 248056,
|
| 7 |
+
"model_type": "qwen3_5_moe",
|
| 8 |
+
"quantization_config": {
|
| 9 |
+
"config_groups": {
|
| 10 |
+
"group_0": {
|
| 11 |
+
"format": "float-quantized",
|
| 12 |
+
"input_activations": {
|
| 13 |
+
"actorder": null,
|
| 14 |
+
"block_structure": null,
|
| 15 |
+
"dynamic": true,
|
| 16 |
+
"group_size": null,
|
| 17 |
+
"num_bits": 8,
|
| 18 |
+
"observer": null,
|
| 19 |
+
"observer_kwargs": {},
|
| 20 |
+
"scale_dtype": null,
|
| 21 |
+
"strategy": "token",
|
| 22 |
+
"symmetric": true,
|
| 23 |
+
"type": "float",
|
| 24 |
+
"zp_dtype": null
|
| 25 |
+
},
|
| 26 |
+
"output_activations": null,
|
| 27 |
+
"targets": [
|
| 28 |
+
"Linear"
|
| 29 |
+
],
|
| 30 |
+
"weights": {
|
| 31 |
+
"actorder": null,
|
| 32 |
+
"block_structure": null,
|
| 33 |
+
"dynamic": false,
|
| 34 |
+
"group_size": null,
|
| 35 |
+
"num_bits": 8,
|
| 36 |
+
"observer": "memoryless_minmax",
|
| 37 |
+
"observer_kwargs": {},
|
| 38 |
+
"scale_dtype": null,
|
| 39 |
+
"strategy": "channel",
|
| 40 |
+
"symmetric": true,
|
| 41 |
+
"type": "float",
|
| 42 |
+
"zp_dtype": null
|
| 43 |
+
}
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"format": "float-quantized",
|
| 47 |
+
"global_compression_ratio": null,
|
| 48 |
+
"ignore": [
|
| 49 |
+
"model.visual.blocks.0.attn.qkv",
|
| 50 |
+
"model.visual.blocks.0.attn.proj",
|
| 51 |
+
"model.visual.blocks.0.mlp.linear_fc1",
|
| 52 |
+
"model.visual.blocks.0.mlp.linear_fc2",
|
| 53 |
+
"model.visual.blocks.1.attn.qkv",
|
| 54 |
+
"model.visual.blocks.1.attn.proj",
|
| 55 |
+
"model.visual.blocks.1.mlp.linear_fc1",
|
| 56 |
+
"model.visual.blocks.1.mlp.linear_fc2",
|
| 57 |
+
"model.visual.blocks.2.attn.qkv",
|
| 58 |
+
"model.visual.blocks.2.attn.proj",
|
| 59 |
+
"model.visual.blocks.2.mlp.linear_fc1",
|
| 60 |
+
"model.visual.blocks.2.mlp.linear_fc2",
|
| 61 |
+
"model.visual.blocks.3.attn.qkv",
|
| 62 |
+
"model.visual.blocks.3.attn.proj",
|
| 63 |
+
"model.visual.blocks.3.mlp.linear_fc1",
|
| 64 |
+
"model.visual.blocks.3.mlp.linear_fc2",
|
| 65 |
+
"model.visual.blocks.4.attn.qkv",
|
| 66 |
+
"model.visual.blocks.4.attn.proj",
|
| 67 |
+
"model.visual.blocks.4.mlp.linear_fc1",
|
| 68 |
+
"model.visual.blocks.4.mlp.linear_fc2",
|
| 69 |
+
"model.visual.blocks.5.attn.qkv",
|
| 70 |
+
"model.visual.blocks.5.attn.proj",
|
| 71 |
+
"model.visual.blocks.5.mlp.linear_fc1",
|
| 72 |
+
"model.visual.blocks.5.mlp.linear_fc2",
|
| 73 |
+
"model.visual.blocks.6.attn.qkv",
|
| 74 |
+
"model.visual.blocks.6.attn.proj",
|
| 75 |
+
"model.visual.blocks.6.mlp.linear_fc1",
|
| 76 |
+
"model.visual.blocks.6.mlp.linear_fc2",
|
| 77 |
+
"model.visual.blocks.7.attn.qkv",
|
| 78 |
+
"model.visual.blocks.7.attn.proj",
|
| 79 |
+
"model.visual.blocks.7.mlp.linear_fc1",
|
| 80 |
+
"model.visual.blocks.7.mlp.linear_fc2",
|
| 81 |
+
"model.visual.blocks.8.attn.qkv",
|
| 82 |
+
"model.visual.blocks.8.attn.proj",
|
| 83 |
+
"model.visual.blocks.8.mlp.linear_fc1",
|
| 84 |
+
"model.visual.blocks.8.mlp.linear_fc2",
|
| 85 |
+
"model.visual.blocks.9.attn.qkv",
|
| 86 |
+
"model.visual.blocks.9.attn.proj",
|
| 87 |
+
"model.visual.blocks.9.mlp.linear_fc1",
|
| 88 |
+
"model.visual.blocks.9.mlp.linear_fc2",
|
| 89 |
+
"model.visual.blocks.10.attn.qkv",
|
| 90 |
+
"model.visual.blocks.10.attn.proj",
|
| 91 |
+
"model.visual.blocks.10.mlp.linear_fc1",
|
| 92 |
+
"model.visual.blocks.10.mlp.linear_fc2",
|
| 93 |
+
"model.visual.blocks.11.attn.qkv",
|
| 94 |
+
"model.visual.blocks.11.attn.proj",
|
| 95 |
+
"model.visual.blocks.11.mlp.linear_fc1",
|
| 96 |
+
"model.visual.blocks.11.mlp.linear_fc2",
|
| 97 |
+
"model.visual.blocks.12.attn.qkv",
|
| 98 |
+
"model.visual.blocks.12.attn.proj",
|
| 99 |
+
"model.visual.blocks.12.mlp.linear_fc1",
|
| 100 |
+
"model.visual.blocks.12.mlp.linear_fc2",
|
| 101 |
+
"model.visual.blocks.13.attn.qkv",
|
| 102 |
+
"model.visual.blocks.13.attn.proj",
|
| 103 |
+
"model.visual.blocks.13.mlp.linear_fc1",
|
| 104 |
+
"model.visual.blocks.13.mlp.linear_fc2",
|
| 105 |
+
"model.visual.blocks.14.attn.qkv",
|
| 106 |
+
"model.visual.blocks.14.attn.proj",
|
| 107 |
+
"model.visual.blocks.14.mlp.linear_fc1",
|
| 108 |
+
"model.visual.blocks.14.mlp.linear_fc2",
|
| 109 |
+
"model.visual.blocks.15.attn.qkv",
|
| 110 |
+
"model.visual.blocks.15.attn.proj",
|
| 111 |
+
"model.visual.blocks.15.mlp.linear_fc1",
|
| 112 |
+
"model.visual.blocks.15.mlp.linear_fc2",
|
| 113 |
+
"model.visual.blocks.16.attn.qkv",
|
| 114 |
+
"model.visual.blocks.16.attn.proj",
|
| 115 |
+
"model.visual.blocks.16.mlp.linear_fc1",
|
| 116 |
+
"model.visual.blocks.16.mlp.linear_fc2",
|
| 117 |
+
"model.visual.blocks.17.attn.qkv",
|
| 118 |
+
"model.visual.blocks.17.attn.proj",
|
| 119 |
+
"model.visual.blocks.17.mlp.linear_fc1",
|
| 120 |
+
"model.visual.blocks.17.mlp.linear_fc2",
|
| 121 |
+
"model.visual.blocks.18.attn.qkv",
|
| 122 |
+
"model.visual.blocks.18.attn.proj",
|
| 123 |
+
"model.visual.blocks.18.mlp.linear_fc1",
|
| 124 |
+
"model.visual.blocks.18.mlp.linear_fc2",
|
| 125 |
+
"model.visual.blocks.19.attn.qkv",
|
| 126 |
+
"model.visual.blocks.19.attn.proj",
|
| 127 |
+
"model.visual.blocks.19.mlp.linear_fc1",
|
| 128 |
+
"model.visual.blocks.19.mlp.linear_fc2",
|
| 129 |
+
"model.visual.blocks.20.attn.qkv",
|
| 130 |
+
"model.visual.blocks.20.attn.proj",
|
| 131 |
+
"model.visual.blocks.20.mlp.linear_fc1",
|
| 132 |
+
"model.visual.blocks.20.mlp.linear_fc2",
|
| 133 |
+
"model.visual.blocks.21.attn.qkv",
|
| 134 |
+
"model.visual.blocks.21.attn.proj",
|
| 135 |
+
"model.visual.blocks.21.mlp.linear_fc1",
|
| 136 |
+
"model.visual.blocks.21.mlp.linear_fc2",
|
| 137 |
+
"model.visual.blocks.22.attn.qkv",
|
| 138 |
+
"model.visual.blocks.22.attn.proj",
|
| 139 |
+
"model.visual.blocks.22.mlp.linear_fc1",
|
| 140 |
+
"model.visual.blocks.22.mlp.linear_fc2",
|
| 141 |
+
"model.visual.blocks.23.attn.qkv",
|
| 142 |
+
"model.visual.blocks.23.attn.proj",
|
| 143 |
+
"model.visual.blocks.23.mlp.linear_fc1",
|
| 144 |
+
"model.visual.blocks.23.mlp.linear_fc2",
|
| 145 |
+
"model.visual.blocks.24.attn.qkv",
|
| 146 |
+
"model.visual.blocks.24.attn.proj",
|
| 147 |
+
"model.visual.blocks.24.mlp.linear_fc1",
|
| 148 |
+
"model.visual.blocks.24.mlp.linear_fc2",
|
| 149 |
+
"model.visual.blocks.25.attn.qkv",
|
| 150 |
+
"model.visual.blocks.25.attn.proj",
|
| 151 |
+
"model.visual.blocks.25.mlp.linear_fc1",
|
| 152 |
+
"model.visual.blocks.25.mlp.linear_fc2",
|
| 153 |
+
"model.visual.blocks.26.attn.qkv",
|
| 154 |
+
"model.visual.blocks.26.attn.proj",
|
| 155 |
+
"model.visual.blocks.26.mlp.linear_fc1",
|
| 156 |
+
"model.visual.blocks.26.mlp.linear_fc2",
|
| 157 |
+
"model.visual.merger.linear_fc1",
|
| 158 |
+
"model.visual.merger.linear_fc2",
|
| 159 |
+
"model.language_model.layers.0.linear_attn.out_proj",
|
| 160 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv",
|
| 161 |
+
"model.language_model.layers.0.linear_attn.in_proj_z",
|
| 162 |
+
"model.language_model.layers.0.linear_attn.in_proj_b",
|
| 163 |
+
"model.language_model.layers.0.linear_attn.in_proj_a",
|
| 164 |
+
"model.language_model.layers.0.mlp.shared_expert_gate",
|
| 165 |
+
"model.language_model.layers.1.linear_attn.out_proj",
|
| 166 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv",
|
| 167 |
+
"model.language_model.layers.1.linear_attn.in_proj_z",
|
| 168 |
+
"model.language_model.layers.1.linear_attn.in_proj_b",
|
| 169 |
+
"model.language_model.layers.1.linear_attn.in_proj_a",
|
| 170 |
+
"model.language_model.layers.1.mlp.shared_expert_gate",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.out_proj",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_z",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_b",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.in_proj_a",
|
| 176 |
+
"model.language_model.layers.2.mlp.shared_expert_gate",
|
| 177 |
+
"model.language_model.layers.3.mlp.shared_expert_gate",
|
| 178 |
+
"model.language_model.layers.4.linear_attn.out_proj",
|
| 179 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv",
|
| 180 |
+
"model.language_model.layers.4.linear_attn.in_proj_z",
|
| 181 |
+
"model.language_model.layers.4.linear_attn.in_proj_b",
|
| 182 |
+
"model.language_model.layers.4.linear_attn.in_proj_a",
|
| 183 |
+
"model.language_model.layers.4.mlp.shared_expert_gate",
|
| 184 |
+
"model.language_model.layers.5.linear_attn.out_proj",
|
| 185 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv",
|
| 186 |
+
"model.language_model.layers.5.linear_attn.in_proj_z",
|
| 187 |
+
"model.language_model.layers.5.linear_attn.in_proj_b",
|
| 188 |
+
"model.language_model.layers.5.linear_attn.in_proj_a",
|
| 189 |
+
"model.language_model.layers.5.mlp.shared_expert_gate",
|
| 190 |
+
"model.language_model.layers.6.linear_attn.out_proj",
|
| 191 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv",
|
| 192 |
+
"model.language_model.layers.6.linear_attn.in_proj_z",
|
| 193 |
+
"model.language_model.layers.6.linear_attn.in_proj_b",
|
| 194 |
+
"model.language_model.layers.6.linear_attn.in_proj_a",
|
| 195 |
+
"model.language_model.layers.6.mlp.shared_expert_gate",
|
| 196 |
+
"model.language_model.layers.7.mlp.shared_expert_gate",
|
| 197 |
+
"model.language_model.layers.8.linear_attn.out_proj",
|
| 198 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv",
|
| 199 |
+
"model.language_model.layers.8.linear_attn.in_proj_z",
|
| 200 |
+
"model.language_model.layers.8.linear_attn.in_proj_b",
|
| 201 |
+
"model.language_model.layers.8.linear_attn.in_proj_a",
|
| 202 |
+
"model.language_model.layers.8.mlp.shared_expert_gate",
|
| 203 |
+
"model.language_model.layers.9.linear_attn.out_proj",
|
| 204 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv",
|
| 205 |
+
"model.language_model.layers.9.linear_attn.in_proj_z",
|
| 206 |
+
"model.language_model.layers.9.linear_attn.in_proj_b",
|
| 207 |
+
"model.language_model.layers.9.linear_attn.in_proj_a",
|
| 208 |
+
"model.language_model.layers.9.mlp.shared_expert_gate",
|
| 209 |
+
"model.language_model.layers.10.linear_attn.out_proj",
|
| 210 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv",
|
| 211 |
+
"model.language_model.layers.10.linear_attn.in_proj_z",
|
| 212 |
+
"model.language_model.layers.10.linear_attn.in_proj_b",
|
| 213 |
+
"model.language_model.layers.10.linear_attn.in_proj_a",
|
| 214 |
+
"model.language_model.layers.10.mlp.shared_expert_gate",
|
| 215 |
+
"model.language_model.layers.11.mlp.shared_expert_gate",
|
| 216 |
+
"model.language_model.layers.12.linear_attn.out_proj",
|
| 217 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv",
|
| 218 |
+
"model.language_model.layers.12.linear_attn.in_proj_z",
|
| 219 |
+
"model.language_model.layers.12.linear_attn.in_proj_b",
|
| 220 |
+
"model.language_model.layers.12.linear_attn.in_proj_a",
|
| 221 |
+
"model.language_model.layers.12.mlp.shared_expert_gate",
|
| 222 |
+
"model.language_model.layers.13.linear_attn.out_proj",
|
| 223 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv",
|
| 224 |
+
"model.language_model.layers.13.linear_attn.in_proj_z",
|
| 225 |
+
"model.language_model.layers.13.linear_attn.in_proj_b",
|
| 226 |
+
"model.language_model.layers.13.linear_attn.in_proj_a",
|
| 227 |
+
"model.language_model.layers.13.mlp.shared_expert_gate",
|
| 228 |
+
"model.language_model.layers.14.linear_attn.out_proj",
|
| 229 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv",
|
| 230 |
+
"model.language_model.layers.14.linear_attn.in_proj_z",
|
| 231 |
+
"model.language_model.layers.14.linear_attn.in_proj_b",
|
| 232 |
+
"model.language_model.layers.14.linear_attn.in_proj_a",
|
| 233 |
+
"model.language_model.layers.14.mlp.shared_expert_gate",
|
| 234 |
+
"model.language_model.layers.15.mlp.shared_expert_gate",
|
| 235 |
+
"model.language_model.layers.16.linear_attn.out_proj",
|
| 236 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv",
|
| 237 |
+
"model.language_model.layers.16.linear_attn.in_proj_z",
|
| 238 |
+
"model.language_model.layers.16.linear_attn.in_proj_b",
|
| 239 |
+
"model.language_model.layers.16.linear_attn.in_proj_a",
|
| 240 |
+
"model.language_model.layers.16.mlp.shared_expert_gate",
|
| 241 |
+
"model.language_model.layers.17.linear_attn.out_proj",
|
| 242 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv",
|
| 243 |
+
"model.language_model.layers.17.linear_attn.in_proj_z",
|
| 244 |
+
"model.language_model.layers.17.linear_attn.in_proj_b",
|
| 245 |
+
"model.language_model.layers.17.linear_attn.in_proj_a",
|
| 246 |
+
"model.language_model.layers.17.mlp.shared_expert_gate",
|
| 247 |
+
"model.language_model.layers.18.linear_attn.out_proj",
|
| 248 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv",
|
| 249 |
+
"model.language_model.layers.18.linear_attn.in_proj_z",
|
| 250 |
+
"model.language_model.layers.18.linear_attn.in_proj_b",
|
| 251 |
+
"model.language_model.layers.18.linear_attn.in_proj_a",
|
| 252 |
+
"model.language_model.layers.18.mlp.shared_expert_gate",
|
| 253 |
+
"model.language_model.layers.19.mlp.shared_expert_gate",
|
| 254 |
+
"model.language_model.layers.20.linear_attn.out_proj",
|
| 255 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv",
|
| 256 |
+
"model.language_model.layers.20.linear_attn.in_proj_z",
|
| 257 |
+
"model.language_model.layers.20.linear_attn.in_proj_b",
|
| 258 |
+
"model.language_model.layers.20.linear_attn.in_proj_a",
|
| 259 |
+
"model.language_model.layers.20.mlp.shared_expert_gate",
|
| 260 |
+
"model.language_model.layers.21.linear_attn.out_proj",
|
| 261 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv",
|
| 262 |
+
"model.language_model.layers.21.linear_attn.in_proj_z",
|
| 263 |
+
"model.language_model.layers.21.linear_attn.in_proj_b",
|
| 264 |
+
"model.language_model.layers.21.linear_attn.in_proj_a",
|
| 265 |
+
"model.language_model.layers.21.mlp.shared_expert_gate",
|
| 266 |
+
"model.language_model.layers.22.linear_attn.out_proj",
|
| 267 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv",
|
| 268 |
+
"model.language_model.layers.22.linear_attn.in_proj_z",
|
| 269 |
+
"model.language_model.layers.22.linear_attn.in_proj_b",
|
| 270 |
+
"model.language_model.layers.22.linear_attn.in_proj_a",
|
| 271 |
+
"model.language_model.layers.22.mlp.shared_expert_gate",
|
| 272 |
+
"model.language_model.layers.23.mlp.shared_expert_gate",
|
| 273 |
+
"model.language_model.layers.24.linear_attn.out_proj",
|
| 274 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv",
|
| 275 |
+
"model.language_model.layers.24.linear_attn.in_proj_z",
|
| 276 |
+
"model.language_model.layers.24.linear_attn.in_proj_b",
|
| 277 |
+
"model.language_model.layers.24.linear_attn.in_proj_a",
|
| 278 |
+
"model.language_model.layers.24.mlp.shared_expert_gate",
|
| 279 |
+
"model.language_model.layers.25.linear_attn.out_proj",
|
| 280 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv",
|
| 281 |
+
"model.language_model.layers.25.linear_attn.in_proj_z",
|
| 282 |
+
"model.language_model.layers.25.linear_attn.in_proj_b",
|
| 283 |
+
"model.language_model.layers.25.linear_attn.in_proj_a",
|
| 284 |
+
"model.language_model.layers.25.mlp.shared_expert_gate",
|
| 285 |
+
"model.language_model.layers.26.linear_attn.out_proj",
|
| 286 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv",
|
| 287 |
+
"model.language_model.layers.26.linear_attn.in_proj_z",
|
| 288 |
+
"model.language_model.layers.26.linear_attn.in_proj_b",
|
| 289 |
+
"model.language_model.layers.26.linear_attn.in_proj_a",
|
| 290 |
+
"model.language_model.layers.26.mlp.shared_expert_gate",
|
| 291 |
+
"model.language_model.layers.27.mlp.shared_expert_gate",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.out_proj",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_z",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.in_proj_b",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.in_proj_a",
|
| 297 |
+
"model.language_model.layers.28.mlp.shared_expert_gate",
|
| 298 |
+
"model.language_model.layers.29.linear_attn.out_proj",
|
| 299 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv",
|
| 300 |
+
"model.language_model.layers.29.linear_attn.in_proj_z",
|
| 301 |
+
"model.language_model.layers.29.linear_attn.in_proj_b",
|
| 302 |
+
"model.language_model.layers.29.linear_attn.in_proj_a",
|
| 303 |
+
"model.language_model.layers.29.mlp.shared_expert_gate",
|
| 304 |
+
"model.language_model.layers.30.linear_attn.out_proj",
|
| 305 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv",
|
| 306 |
+
"model.language_model.layers.30.linear_attn.in_proj_z",
|
| 307 |
+
"model.language_model.layers.30.linear_attn.in_proj_b",
|
| 308 |
+
"model.language_model.layers.30.linear_attn.in_proj_a",
|
| 309 |
+
"model.language_model.layers.30.mlp.shared_expert_gate",
|
| 310 |
+
"model.language_model.layers.31.mlp.shared_expert_gate",
|
| 311 |
+
"model.language_model.layers.32.linear_attn.out_proj",
|
| 312 |
+
"model.language_model.layers.32.linear_attn.in_proj_qkv",
|
| 313 |
+
"model.language_model.layers.32.linear_attn.in_proj_z",
|
| 314 |
+
"model.language_model.layers.32.linear_attn.in_proj_b",
|
| 315 |
+
"model.language_model.layers.32.linear_attn.in_proj_a",
|
| 316 |
+
"model.language_model.layers.32.mlp.shared_expert_gate",
|
| 317 |
+
"model.language_model.layers.33.linear_attn.out_proj",
|
| 318 |
+
"model.language_model.layers.33.linear_attn.in_proj_qkv",
|
| 319 |
+
"model.language_model.layers.33.linear_attn.in_proj_z",
|
| 320 |
+
"model.language_model.layers.33.linear_attn.in_proj_b",
|
| 321 |
+
"model.language_model.layers.33.linear_attn.in_proj_a",
|
| 322 |
+
"model.language_model.layers.33.mlp.shared_expert_gate",
|
| 323 |
+
"model.language_model.layers.34.linear_attn.out_proj",
|
| 324 |
+
"model.language_model.layers.34.linear_attn.in_proj_qkv",
|
| 325 |
+
"model.language_model.layers.34.linear_attn.in_proj_z",
|
| 326 |
+
"model.language_model.layers.34.linear_attn.in_proj_b",
|
| 327 |
+
"model.language_model.layers.34.linear_attn.in_proj_a",
|
| 328 |
+
"model.language_model.layers.34.mlp.shared_expert_gate",
|
| 329 |
+
"model.language_model.layers.35.mlp.shared_expert_gate",
|
| 330 |
+
"model.language_model.layers.36.linear_attn.out_proj",
|
| 331 |
+
"model.language_model.layers.36.linear_attn.in_proj_qkv",
|
| 332 |
+
"model.language_model.layers.36.linear_attn.in_proj_z",
|
| 333 |
+
"model.language_model.layers.36.linear_attn.in_proj_b",
|
| 334 |
+
"model.language_model.layers.36.linear_attn.in_proj_a",
|
| 335 |
+
"model.language_model.layers.36.mlp.shared_expert_gate",
|
| 336 |
+
"model.language_model.layers.37.linear_attn.out_proj",
|
| 337 |
+
"model.language_model.layers.37.linear_attn.in_proj_qkv",
|
| 338 |
+
"model.language_model.layers.37.linear_attn.in_proj_z",
|
| 339 |
+
"model.language_model.layers.37.linear_attn.in_proj_b",
|
| 340 |
+
"model.language_model.layers.37.linear_attn.in_proj_a",
|
| 341 |
+
"model.language_model.layers.37.mlp.shared_expert_gate",
|
| 342 |
+
"model.language_model.layers.38.linear_attn.out_proj",
|
| 343 |
+
"model.language_model.layers.38.linear_attn.in_proj_qkv",
|
| 344 |
+
"model.language_model.layers.38.linear_attn.in_proj_z",
|
| 345 |
+
"model.language_model.layers.38.linear_attn.in_proj_b",
|
| 346 |
+
"model.language_model.layers.38.linear_attn.in_proj_a",
|
| 347 |
+
"model.language_model.layers.38.mlp.shared_expert_gate",
|
| 348 |
+
"model.language_model.layers.39.mlp.shared_expert_gate",
|
| 349 |
+
"model.language_model.layers.40.linear_attn.out_proj",
|
| 350 |
+
"model.language_model.layers.40.linear_attn.in_proj_qkv",
|
| 351 |
+
"model.language_model.layers.40.linear_attn.in_proj_z",
|
| 352 |
+
"model.language_model.layers.40.linear_attn.in_proj_b",
|
| 353 |
+
"model.language_model.layers.40.linear_attn.in_proj_a",
|
| 354 |
+
"model.language_model.layers.40.mlp.shared_expert_gate",
|
| 355 |
+
"model.language_model.layers.41.linear_attn.out_proj",
|
| 356 |
+
"model.language_model.layers.41.linear_attn.in_proj_qkv",
|
| 357 |
+
"model.language_model.layers.41.linear_attn.in_proj_z",
|
| 358 |
+
"model.language_model.layers.41.linear_attn.in_proj_b",
|
| 359 |
+
"model.language_model.layers.41.linear_attn.in_proj_a",
|
| 360 |
+
"model.language_model.layers.41.mlp.shared_expert_gate",
|
| 361 |
+
"model.language_model.layers.42.linear_attn.out_proj",
|
| 362 |
+
"model.language_model.layers.42.linear_attn.in_proj_qkv",
|
| 363 |
+
"model.language_model.layers.42.linear_attn.in_proj_z",
|
| 364 |
+
"model.language_model.layers.42.linear_attn.in_proj_b",
|
| 365 |
+
"model.language_model.layers.42.linear_attn.in_proj_a",
|
| 366 |
+
"model.language_model.layers.42.mlp.shared_expert_gate",
|
| 367 |
+
"model.language_model.layers.43.mlp.shared_expert_gate",
|
| 368 |
+
"model.language_model.layers.44.linear_attn.out_proj",
|
| 369 |
+
"model.language_model.layers.44.linear_attn.in_proj_qkv",
|
| 370 |
+
"model.language_model.layers.44.linear_attn.in_proj_z",
|
| 371 |
+
"model.language_model.layers.44.linear_attn.in_proj_b",
|
| 372 |
+
"model.language_model.layers.44.linear_attn.in_proj_a",
|
| 373 |
+
"model.language_model.layers.44.mlp.shared_expert_gate",
|
| 374 |
+
"model.language_model.layers.45.linear_attn.out_proj",
|
| 375 |
+
"model.language_model.layers.45.linear_attn.in_proj_qkv",
|
| 376 |
+
"model.language_model.layers.45.linear_attn.in_proj_z",
|
| 377 |
+
"model.language_model.layers.45.linear_attn.in_proj_b",
|
| 378 |
+
"model.language_model.layers.45.linear_attn.in_proj_a",
|
| 379 |
+
"model.language_model.layers.45.mlp.shared_expert_gate",
|
| 380 |
+
"model.language_model.layers.46.linear_attn.out_proj",
|
| 381 |
+
"model.language_model.layers.46.linear_attn.in_proj_qkv",
|
| 382 |
+
"model.language_model.layers.46.linear_attn.in_proj_z",
|
| 383 |
+
"model.language_model.layers.46.linear_attn.in_proj_b",
|
| 384 |
+
"model.language_model.layers.46.linear_attn.in_proj_a",
|
| 385 |
+
"model.language_model.layers.46.mlp.shared_expert_gate",
|
| 386 |
+
"model.language_model.layers.47.mlp.shared_expert_gate",
|
| 387 |
+
"model.language_model.layers.48.linear_attn.out_proj",
|
| 388 |
+
"model.language_model.layers.48.linear_attn.in_proj_qkv",
|
| 389 |
+
"model.language_model.layers.48.linear_attn.in_proj_z",
|
| 390 |
+
"model.language_model.layers.48.linear_attn.in_proj_b",
|
| 391 |
+
"model.language_model.layers.48.linear_attn.in_proj_a",
|
| 392 |
+
"model.language_model.layers.48.mlp.shared_expert_gate",
|
| 393 |
+
"model.language_model.layers.49.linear_attn.out_proj",
|
| 394 |
+
"model.language_model.layers.49.linear_attn.in_proj_qkv",
|
| 395 |
+
"model.language_model.layers.49.linear_attn.in_proj_z",
|
| 396 |
+
"model.language_model.layers.49.linear_attn.in_proj_b",
|
| 397 |
+
"model.language_model.layers.49.linear_attn.in_proj_a",
|
| 398 |
+
"model.language_model.layers.49.mlp.shared_expert_gate",
|
| 399 |
+
"model.language_model.layers.50.linear_attn.out_proj",
|
| 400 |
+
"model.language_model.layers.50.linear_attn.in_proj_qkv",
|
| 401 |
+
"model.language_model.layers.50.linear_attn.in_proj_z",
|
| 402 |
+
"model.language_model.layers.50.linear_attn.in_proj_b",
|
| 403 |
+
"model.language_model.layers.50.linear_attn.in_proj_a",
|
| 404 |
+
"model.language_model.layers.50.mlp.shared_expert_gate",
|
| 405 |
+
"model.language_model.layers.51.mlp.shared_expert_gate",
|
| 406 |
+
"model.language_model.layers.52.linear_attn.out_proj",
|
| 407 |
+
"model.language_model.layers.52.linear_attn.in_proj_qkv",
|
| 408 |
+
"model.language_model.layers.52.linear_attn.in_proj_z",
|
| 409 |
+
"model.language_model.layers.52.linear_attn.in_proj_b",
|
| 410 |
+
"model.language_model.layers.52.linear_attn.in_proj_a",
|
| 411 |
+
"model.language_model.layers.52.mlp.shared_expert_gate",
|
| 412 |
+
"model.language_model.layers.53.linear_attn.out_proj",
|
| 413 |
+
"model.language_model.layers.53.linear_attn.in_proj_qkv",
|
| 414 |
+
"model.language_model.layers.53.linear_attn.in_proj_z",
|
| 415 |
+
"model.language_model.layers.53.linear_attn.in_proj_b",
|
| 416 |
+
"model.language_model.layers.53.linear_attn.in_proj_a",
|
| 417 |
+
"model.language_model.layers.53.mlp.shared_expert_gate",
|
| 418 |
+
"model.language_model.layers.54.linear_attn.out_proj",
|
| 419 |
+
"model.language_model.layers.54.linear_attn.in_proj_qkv",
|
| 420 |
+
"model.language_model.layers.54.linear_attn.in_proj_z",
|
| 421 |
+
"model.language_model.layers.54.linear_attn.in_proj_b",
|
| 422 |
+
"model.language_model.layers.54.linear_attn.in_proj_a",
|
| 423 |
+
"model.language_model.layers.54.mlp.shared_expert_gate",
|
| 424 |
+
"model.language_model.layers.55.mlp.shared_expert_gate",
|
| 425 |
+
"model.language_model.layers.56.linear_attn.out_proj",
|
| 426 |
+
"model.language_model.layers.56.linear_attn.in_proj_qkv",
|
| 427 |
+
"model.language_model.layers.56.linear_attn.in_proj_z",
|
| 428 |
+
"model.language_model.layers.56.linear_attn.in_proj_b",
|
| 429 |
+
"model.language_model.layers.56.linear_attn.in_proj_a",
|
| 430 |
+
"model.language_model.layers.56.mlp.shared_expert_gate",
|
| 431 |
+
"model.language_model.layers.57.linear_attn.out_proj",
|
| 432 |
+
"model.language_model.layers.57.linear_attn.in_proj_qkv",
|
| 433 |
+
"model.language_model.layers.57.linear_attn.in_proj_z",
|
| 434 |
+
"model.language_model.layers.57.linear_attn.in_proj_b",
|
| 435 |
+
"model.language_model.layers.57.linear_attn.in_proj_a",
|
| 436 |
+
"model.language_model.layers.57.mlp.shared_expert_gate",
|
| 437 |
+
"model.language_model.layers.58.linear_attn.out_proj",
|
| 438 |
+
"model.language_model.layers.58.linear_attn.in_proj_qkv",
|
| 439 |
+
"model.language_model.layers.58.linear_attn.in_proj_z",
|
| 440 |
+
"model.language_model.layers.58.linear_attn.in_proj_b",
|
| 441 |
+
"model.language_model.layers.58.linear_attn.in_proj_a",
|
| 442 |
+
"model.language_model.layers.58.mlp.shared_expert_gate",
|
| 443 |
+
"model.language_model.layers.59.mlp.shared_expert_gate",
|
| 444 |
+
"model.language_model.layers.0.mlp.gate",
|
| 445 |
+
"model.language_model.layers.1.mlp.gate",
|
| 446 |
+
"model.language_model.layers.2.mlp.gate",
|
| 447 |
+
"model.language_model.layers.3.mlp.gate",
|
| 448 |
+
"model.language_model.layers.4.mlp.gate",
|
| 449 |
+
"model.language_model.layers.5.mlp.gate",
|
| 450 |
+
"model.language_model.layers.6.mlp.gate",
|
| 451 |
+
"model.language_model.layers.7.mlp.gate",
|
| 452 |
+
"model.language_model.layers.8.mlp.gate",
|
| 453 |
+
"model.language_model.layers.9.mlp.gate",
|
| 454 |
+
"model.language_model.layers.10.mlp.gate",
|
| 455 |
+
"model.language_model.layers.11.mlp.gate",
|
| 456 |
+
"model.language_model.layers.12.mlp.gate",
|
| 457 |
+
"model.language_model.layers.13.mlp.gate",
|
| 458 |
+
"model.language_model.layers.14.mlp.gate",
|
| 459 |
+
"model.language_model.layers.15.mlp.gate",
|
| 460 |
+
"model.language_model.layers.16.mlp.gate",
|
| 461 |
+
"model.language_model.layers.17.mlp.gate",
|
| 462 |
+
"model.language_model.layers.18.mlp.gate",
|
| 463 |
+
"model.language_model.layers.19.mlp.gate",
|
| 464 |
+
"model.language_model.layers.20.mlp.gate",
|
| 465 |
+
"model.language_model.layers.21.mlp.gate",
|
| 466 |
+
"model.language_model.layers.22.mlp.gate",
|
| 467 |
+
"model.language_model.layers.23.mlp.gate",
|
| 468 |
+
"model.language_model.layers.24.mlp.gate",
|
| 469 |
+
"model.language_model.layers.25.mlp.gate",
|
| 470 |
+
"model.language_model.layers.26.mlp.gate",
|
| 471 |
+
"model.language_model.layers.27.mlp.gate",
|
| 472 |
+
"model.language_model.layers.28.mlp.gate",
|
| 473 |
+
"model.language_model.layers.29.mlp.gate",
|
| 474 |
+
"model.language_model.layers.30.mlp.gate",
|
| 475 |
+
"model.language_model.layers.31.mlp.gate",
|
| 476 |
+
"model.language_model.layers.32.mlp.gate",
|
| 477 |
+
"model.language_model.layers.33.mlp.gate",
|
| 478 |
+
"model.language_model.layers.34.mlp.gate",
|
| 479 |
+
"model.language_model.layers.35.mlp.gate",
|
| 480 |
+
"model.language_model.layers.36.mlp.gate",
|
| 481 |
+
"model.language_model.layers.37.mlp.gate",
|
| 482 |
+
"model.language_model.layers.38.mlp.gate",
|
| 483 |
+
"model.language_model.layers.39.mlp.gate",
|
| 484 |
+
"model.language_model.layers.40.mlp.gate",
|
| 485 |
+
"model.language_model.layers.41.mlp.gate",
|
| 486 |
+
"model.language_model.layers.42.mlp.gate",
|
| 487 |
+
"model.language_model.layers.43.mlp.gate",
|
| 488 |
+
"model.language_model.layers.44.mlp.gate",
|
| 489 |
+
"model.language_model.layers.45.mlp.gate",
|
| 490 |
+
"model.language_model.layers.46.mlp.gate",
|
| 491 |
+
"model.language_model.layers.47.mlp.gate",
|
| 492 |
+
"model.language_model.layers.48.mlp.gate",
|
| 493 |
+
"model.language_model.layers.49.mlp.gate",
|
| 494 |
+
"model.language_model.layers.50.mlp.gate",
|
| 495 |
+
"model.language_model.layers.51.mlp.gate",
|
| 496 |
+
"model.language_model.layers.52.mlp.gate",
|
| 497 |
+
"model.language_model.layers.53.mlp.gate",
|
| 498 |
+
"model.language_model.layers.54.mlp.gate",
|
| 499 |
+
"model.language_model.layers.55.mlp.gate",
|
| 500 |
+
"model.language_model.layers.56.mlp.gate",
|
| 501 |
+
"model.language_model.layers.57.mlp.gate",
|
| 502 |
+
"model.language_model.layers.58.mlp.gate",
|
| 503 |
+
"model.language_model.layers.59.mlp.gate",
|
| 504 |
+
"lm_head"
|
| 505 |
+
],
|
| 506 |
+
"kv_cache_scheme": null,
|
| 507 |
+
"quant_method": "compressed-tensors",
|
| 508 |
+
"quantization_status": "compressed",
|
| 509 |
+
"sparsity_config": {},
|
| 510 |
+
"transform_config": {},
|
| 511 |
+
"version": "0.13.1.a20260212"
|
| 512 |
+
},
|
| 513 |
+
"text_config": {
|
| 514 |
+
"attention_bias": false,
|
| 515 |
+
"attention_dropout": 0.0,
|
| 516 |
+
"attn_output_gate": true,
|
| 517 |
+
"bos_token_id": null,
|
| 518 |
+
"dtype": "bfloat16",
|
| 519 |
+
"eos_token_id": 248044,
|
| 520 |
+
"full_attention_interval": 4,
|
| 521 |
+
"head_dim": 256,
|
| 522 |
+
"hidden_act": "silu",
|
| 523 |
+
"hidden_size": 4096,
|
| 524 |
+
"initializer_range": 0.02,
|
| 525 |
+
"layer_types": [
|
| 526 |
+
"linear_attention",
|
| 527 |
+
"linear_attention",
|
| 528 |
+
"linear_attention",
|
| 529 |
+
"full_attention",
|
| 530 |
+
"linear_attention",
|
| 531 |
+
"linear_attention",
|
| 532 |
+
"linear_attention",
|
| 533 |
+
"full_attention",
|
| 534 |
+
"linear_attention",
|
| 535 |
+
"linear_attention",
|
| 536 |
+
"linear_attention",
|
| 537 |
+
"full_attention",
|
| 538 |
+
"linear_attention",
|
| 539 |
+
"linear_attention",
|
| 540 |
+
"linear_attention",
|
| 541 |
+
"full_attention",
|
| 542 |
+
"linear_attention",
|
| 543 |
+
"linear_attention",
|
| 544 |
+
"linear_attention",
|
| 545 |
+
"full_attention",
|
| 546 |
+
"linear_attention",
|
| 547 |
+
"linear_attention",
|
| 548 |
+
"linear_attention",
|
| 549 |
+
"full_attention",
|
| 550 |
+
"linear_attention",
|
| 551 |
+
"linear_attention",
|
| 552 |
+
"linear_attention",
|
| 553 |
+
"full_attention",
|
| 554 |
+
"linear_attention",
|
| 555 |
+
"linear_attention",
|
| 556 |
+
"linear_attention",
|
| 557 |
+
"full_attention",
|
| 558 |
+
"linear_attention",
|
| 559 |
+
"linear_attention",
|
| 560 |
+
"linear_attention",
|
| 561 |
+
"full_attention",
|
| 562 |
+
"linear_attention",
|
| 563 |
+
"linear_attention",
|
| 564 |
+
"linear_attention",
|
| 565 |
+
"full_attention",
|
| 566 |
+
"linear_attention",
|
| 567 |
+
"linear_attention",
|
| 568 |
+
"linear_attention",
|
| 569 |
+
"full_attention",
|
| 570 |
+
"linear_attention",
|
| 571 |
+
"linear_attention",
|
| 572 |
+
"linear_attention",
|
| 573 |
+
"full_attention",
|
| 574 |
+
"linear_attention",
|
| 575 |
+
"linear_attention",
|
| 576 |
+
"linear_attention",
|
| 577 |
+
"full_attention",
|
| 578 |
+
"linear_attention",
|
| 579 |
+
"linear_attention",
|
| 580 |
+
"linear_attention",
|
| 581 |
+
"full_attention",
|
| 582 |
+
"linear_attention",
|
| 583 |
+
"linear_attention",
|
| 584 |
+
"linear_attention",
|
| 585 |
+
"full_attention"
|
| 586 |
+
],
|
| 587 |
+
"linear_conv_kernel_dim": 4,
|
| 588 |
+
"linear_key_head_dim": 128,
|
| 589 |
+
"linear_num_key_heads": 16,
|
| 590 |
+
"linear_num_value_heads": 64,
|
| 591 |
+
"linear_value_head_dim": 128,
|
| 592 |
+
"mamba_ssm_dtype": "float32",
|
| 593 |
+
"max_position_embeddings": 262144,
|
| 594 |
+
"mlp_only_layers": [],
|
| 595 |
+
"model_type": "qwen3_5_moe_text",
|
| 596 |
+
"moe_intermediate_size": 1024,
|
| 597 |
+
"mtp_num_hidden_layers": 1,
|
| 598 |
+
"mtp_use_dedicated_embeddings": false,
|
| 599 |
+
"num_attention_heads": 32,
|
| 600 |
+
"num_experts": 512,
|
| 601 |
+
"num_experts_per_tok": 10,
|
| 602 |
+
"num_hidden_layers": 60,
|
| 603 |
+
"num_key_value_heads": 2,
|
| 604 |
+
"output_router_logits": false,
|
| 605 |
+
"pad_token_id": null,
|
| 606 |
+
"partial_rotary_factor": 0.25,
|
| 607 |
+
"rms_norm_eps": 1e-06,
|
| 608 |
+
"rope_parameters": {
|
| 609 |
+
"mrope_interleaved": true,
|
| 610 |
+
"mrope_section": [
|
| 611 |
+
11,
|
| 612 |
+
11,
|
| 613 |
+
10
|
| 614 |
+
],
|
| 615 |
+
"partial_rotary_factor": 0.25,
|
| 616 |
+
"rope_theta": 10000000,
|
| 617 |
+
"rope_type": "default"
|
| 618 |
+
},
|
| 619 |
+
"router_aux_loss_coef": 0.001,
|
| 620 |
+
"shared_expert_intermediate_size": 1024,
|
| 621 |
+
"tie_word_embeddings": false,
|
| 622 |
+
"use_cache": true,
|
| 623 |
+
"vocab_size": 248320
|
| 624 |
+
},
|
| 625 |
+
"tie_word_embeddings": false,
|
| 626 |
+
"transformers_version": "5.2.0",
|
| 627 |
+
"video_token_id": 248057,
|
| 628 |
+
"vision_config": {
|
| 629 |
+
"deepstack_visual_indexes": [],
|
| 630 |
+
"depth": 27,
|
| 631 |
+
"dtype": "bfloat16",
|
| 632 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 633 |
+
"hidden_size": 1152,
|
| 634 |
+
"in_channels": 3,
|
| 635 |
+
"initializer_range": 0.02,
|
| 636 |
+
"intermediate_size": 4304,
|
| 637 |
+
"model_type": "qwen3_5_moe",
|
| 638 |
+
"num_heads": 16,
|
| 639 |
+
"num_position_embeddings": 2304,
|
| 640 |
+
"out_hidden_size": 4096,
|
| 641 |
+
"patch_size": 16,
|
| 642 |
+
"spatial_merge_size": 2,
|
| 643 |
+
"temporal_patch_size": 2
|
| 644 |
+
},
|
| 645 |
+
"vision_end_token_id": 248054,
|
| 646 |
+
"vision_start_token_id": 248053
|
| 647 |
+
}
|
recipe.yaml
CHANGED
|
@@ -3,6 +3,5 @@ default_stage:
|
|
| 3 |
QuantizationModifier:
|
| 4 |
targets: [Linear]
|
| 5 |
ignore: ['re:.*lm_head', 're:visual.*', 're:model.visual.*', 're:.*mlp.gate$', 're:.*embed_tokens$',
|
| 6 |
-
're:.*shared_expert_gate$', 're:.*
|
| 7 |
-
're:.*in_proj_ba$', 're:.*norm$', 're:.*pre_fc_norm_hidden$', 're:.*pre_fc_norm_embedding$']
|
| 8 |
scheme: FP8_DYNAMIC
|
|
|
|
| 3 |
QuantizationModifier:
|
| 4 |
targets: [Linear]
|
| 5 |
ignore: ['re:.*lm_head', 're:visual.*', 're:model.visual.*', 're:.*mlp.gate$', 're:.*embed_tokens$',
|
| 6 |
+
're:.*shared_expert_gate$', 're:.*linear_attn.*']
|
|
|
|
| 7 |
scheme: FP8_DYNAMIC
|