Download quant_config.json from Baranee2007/24b1814-Week08-Compression40-Submission01: direct link, hf CLI and curl.
- Browser
- Download file 49.8 kB
-
https://huggingface.co/Baranee2007/24b1814-Week08-Compression40-Submission01/resolve/main/quant_config.json
- Command line
-
hf download hf://Baranee2007/24b1814-Week08-Compression40-Submission01/quant_config.json
-
curl -L -o quant_config.json https://huggingface.co/Baranee2007/24b1814-Week08-Compression40-Submission01/resolve/main/quant_config.json
49.8 kB
| { | |
| "format": "cs6013-groupwise-int-v1", | |
| "tensors": { | |
| "model.language_model.embed_tokens.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 248320, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.1.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.10.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.11.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.11.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.11.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.12.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.13.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.14.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.15.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.15.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.15.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.16.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.17.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.18.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.19.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.19.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.19.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.2.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.20.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.21.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.22.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.23.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.23.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.23.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.24.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.25.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.29.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.3.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.3.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.3.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.30.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.31.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.31.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.31.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.4.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.5.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.6.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.7.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.7.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.7.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.8.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.9.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "mtp.layers.0.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "mtp.layers.0.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "mtp.layers.0.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.0.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.0.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.0.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.1.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.10.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.10.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.11.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.11.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.11.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.11.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.12.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.13.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.13.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.14.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.14.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.15.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.15.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.15.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.15.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.16.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.17.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.17.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.18.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.18.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.19.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.19.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.19.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.19.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.2.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.20.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.20.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.21.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.21.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.22.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.22.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.23.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.23.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.23.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.23.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.24.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.25.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.25.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.26.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.26.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.26.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.27.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.27.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.27.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.27.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.27.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.27.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.27.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.28.mlp.down_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 9216 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 9216 | |
| }, | |
| "model.language_model.layers.28.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.28.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.29.mlp.gate_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.29.mlp.up_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 9216, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.3.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.3.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.3.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.3.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.30.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.31.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.31.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.31.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.31.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.4.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.5.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.5.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.6.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.6.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.7.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.7.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.7.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.7.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.8.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "model.language_model.layers.9.linear_attn.in_proj_a.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.linear_attn.in_proj_b.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 32, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.linear_attn.in_proj_qkv.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.linear_attn.in_proj_z.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 4096, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "model.language_model.layers.9.linear_attn.out_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "mtp.fc.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 5120 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 5120 | |
| }, | |
| "mtp.layers.0.self_attn.k_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "mtp.layers.0.self_attn.o_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 2560, | |
| 4096 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 4096 | |
| }, | |
| "mtp.layers.0.self_attn.q_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 8192, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| }, | |
| "mtp.layers.0.self_attn.v_proj.weight": { | |
| "bits": 6, | |
| "group_size": 128, | |
| "shape": [ | |
| 1024, | |
| 2560 | |
| ], | |
| "orig_dtype": "bfloat16", | |
| "padded_cols": 2560 | |
| } | |
| } | |
| } |