Download tensor-cache.json from logitsml/drowse-web-qwen35-2b-base-candidate: direct link, hf CLI and curl.
- Browser
- Download file 173 kB
-
https://huggingface.co/logitsml/drowse-web-qwen35-2b-base-candidate/resolve/main/tensor-cache.json
- Command line
-
hf download hf://logitsml/drowse-web-qwen35-2b-base-candidate/tensor-cache.json
-
curl -L -o tensor-cache.json https://huggingface.co/logitsml/drowse-web-qwen35-2b-base-candidate/resolve/main/tensor-cache.json
173 kB
| { | |
| "metadata": { | |
| "ParamSize": 435, | |
| "ParamBytes": 1059316480.0, | |
| "BitsPerParam": 3.726592627263146 | |
| }, | |
| "records": [ | |
| { | |
| "dataPath": "params_shard_0.bin", | |
| "format": "raw-shard", | |
| "nbytes": 254279680, | |
| "records": [ | |
| { | |
| "name": "model.embed_tokens.q_weight", | |
| "shape": [ | |
| 248320, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 254279680, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "422f68032ef81a1dbd977a771b95e7fd" | |
| }, | |
| { | |
| "dataPath": "params_shard_1.bin", | |
| "format": "raw-shard", | |
| "nbytes": 31784960, | |
| "records": [ | |
| { | |
| "name": "model.embed_tokens.q_scale", | |
| "shape": [ | |
| 248320, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 31784960, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "214b6ded9f4b085cef1ad15823704172" | |
| }, | |
| { | |
| "dataPath": "params_shard_2.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.11.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.11.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.11.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.11.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 19660800 | |
| }, | |
| { | |
| "name": "model.layers.1.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.1.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "0da602feee28af6401a6aaa2c77eace3" | |
| }, | |
| { | |
| "dataPath": "params_shard_3.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.1.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.1.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.12.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.12.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "8207c7df5e01917f8c83d8c4fc2947c0" | |
| }, | |
| { | |
| "dataPath": "params_shard_4.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.12.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.12.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.16.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.16.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "3e184bac76a2c9fce27ede8a832fdefc" | |
| }, | |
| { | |
| "dataPath": "params_shard_5.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.16.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.16.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.15.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.15.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "609567a03ecb7ee20e40a047a5662d21" | |
| }, | |
| { | |
| "dataPath": "params_shard_6.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.15.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.15.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.18.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.18.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "9e1f9a2daec0814fe95064f3b6dba6d2" | |
| }, | |
| { | |
| "dataPath": "params_shard_7.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.18.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.18.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.19.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.19.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "d40abda00657776b6d27fd1141d2c7f2" | |
| }, | |
| { | |
| "dataPath": "params_shard_8.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.0.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.0.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.0.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.0.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "607b9fd78c8486f639e78a463b753773" | |
| }, | |
| { | |
| "dataPath": "params_shard_9.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.13.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.13.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.13.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.13.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 19660800 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "365ce463b6e863128da5766fa2fa464a" | |
| }, | |
| { | |
| "dataPath": "params_shard_10.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.10.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.10.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.10.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.10.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "ec488164724c8e24d3daf8704d27e56b" | |
| }, | |
| { | |
| "dataPath": "params_shard_11.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.14.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.14.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.14.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.14.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "d8120f1fb1d0b91ac4370be5f0a226f1" | |
| }, | |
| { | |
| "dataPath": "params_shard_12.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.19.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.19.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.2.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.2.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.2.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.2.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "b3ae8a11f6503536aa410cb7d7c8ddc3" | |
| }, | |
| { | |
| "dataPath": "params_shard_13.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.21.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.21.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.21.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.21.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "b6ba73a9385b3680ac682ca116e438a4" | |
| }, | |
| { | |
| "dataPath": "params_shard_14.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.22.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.22.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.22.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.22.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 19660800 | |
| }, | |
| { | |
| "name": "model.layers.23.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.23.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "37172bfaa484e56c22e3f86e9ec24a0e" | |
| }, | |
| { | |
| "dataPath": "params_shard_15.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.23.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.23.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| }, | |
| { | |
| "name": "model.layers.17.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.17.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "4b94993a286c3778ae8a6e34c8e32ae0" | |
| }, | |
| { | |
| "dataPath": "params_shard_16.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.5.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.5.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "b1e949707e48ce64cbfd714ba796d316" | |
| }, | |
| { | |
| "dataPath": "params_shard_17.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.5.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.5.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| }, | |
| { | |
| "name": "model.layers.20.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.20.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "4d98409431fe2f6e9b3a053ccb99d350" | |
| }, | |
| { | |
| "dataPath": "params_shard_18.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.20.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.20.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.3.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.3.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "77e193769cabac297b88f2fc55227cdc" | |
| }, | |
| { | |
| "dataPath": "params_shard_19.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.3.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.3.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "bf17e41e5958249e024833b1e8b1b644" | |
| }, | |
| { | |
| "dataPath": "params_shard_20.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.7.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.7.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "f5dbcdfbd56269201dfee834ab01cc5f" | |
| }, | |
| { | |
| "dataPath": "params_shard_21.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.17.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.17.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.9.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.9.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "45e10fb1fbe5c4a63ce3279462e77534" | |
| }, | |
| { | |
| "dataPath": "params_shard_22.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.9.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.9.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.8.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.8.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "bb7ac88b4a4abc581b15a06b583842ce" | |
| }, | |
| { | |
| "dataPath": "params_shard_23.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.8.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.8.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.6.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.6.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "1a9f94d8a6f4e20bd568a3e30140302a" | |
| }, | |
| { | |
| "dataPath": "params_shard_24.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.6.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.6.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.7.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.7.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 26738688 | |
| } | |
| ], | |
| "md5sum": "0c850b15de60e52f9ccc29d97cbe5704" | |
| }, | |
| { | |
| "dataPath": "params_shard_25.bin", | |
| "format": "raw-shard", | |
| "nbytes": 21233664, | |
| "records": [ | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.4.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 768 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.4.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 192 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| } | |
| ], | |
| "md5sum": "df425a9a24b04992c343cdfe0e69411d" | |
| }, | |
| { | |
| "dataPath": "params_shard_26.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28311552, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 12288, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 12582912, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.4.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 12288, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1572864, | |
| "byteOffset": 12582912 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 20447232 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 27525120 | |
| } | |
| ], | |
| "md5sum": "5674b78e93c0f633a0d0b2487f397bc8" | |
| }, | |
| { | |
| "dataPath": "params_shard_27.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30670848, | |
| "records": [ | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_qkv.q_weight", | |
| "shape": [ | |
| 6144, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6291456, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_qkv.q_scale", | |
| "shape": [ | |
| 6144, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 12320768 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 12976128 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 18219008 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 18874368 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 24117248 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 24772608 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 30015488 | |
| } | |
| ], | |
| "md5sum": "eb5f523c37750c3fb5a64f098e75d22e" | |
| }, | |
| { | |
| "dataPath": "params_shard_28.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33030144, | |
| "records": [ | |
| { | |
| "name": "model.layers.3.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 5242880 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.c_attn.q_weight", | |
| "shape": [ | |
| 5120, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 5242880, | |
| "byteOffset": 5898240 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.c_attn.q_scale", | |
| "shape": [ | |
| 5120, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 655360, | |
| "byteOffset": 11141120 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 11796480 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 13893632 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 16252928 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 16515072 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 18612224 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 18874368 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 20971520 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 23330816 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 23592960 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 25690112 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 25952256 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28049408 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28311552 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 30408704 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 30670848 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 32768000 | |
| } | |
| ], | |
| "md5sum": "4167692daaf0dbbaca63ca974019b095" | |
| }, | |
| { | |
| "dataPath": "params_shard_29.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33030144, | |
| "records": [ | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 2097152 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2359296 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 4456448 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 4718592 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 6815744 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9175040 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 9437184 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 11534336 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 11796480 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 13893632 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 16252928 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 16515072 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 18612224 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 18874368 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 20971520 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 23330816 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 23592960 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 25690112 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 25952256 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28049408 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28311552 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 30408704 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 30670848 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 32768000 | |
| } | |
| ], | |
| "md5sum": "068a337ba91543451f2a765218ff820c" | |
| }, | |
| { | |
| "dataPath": "params_shard_30.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33030144, | |
| "records": [ | |
| { | |
| "name": "model.layers.20.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 2097152 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2359296 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 4456448 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 4718592 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 6815744 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9175040 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 9437184 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 11534336 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 11796480 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 13893632 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 14155776 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 16252928 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 16515072 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 18612224 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 18874368 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 20971520 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 21233664 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 23330816 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 23592960 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 25690112 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 25952256 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28049408 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28311552 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 30408704 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 30670848 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 32768000 | |
| } | |
| ], | |
| "md5sum": "05655ffa0513017aac9dab09b55eb0e5" | |
| }, | |
| { | |
| "dataPath": "params_shard_31.bin", | |
| "format": "raw-shard", | |
| "nbytes": 13557376, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 2097152 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.out_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2359296 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.out_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 4456448 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 4718592 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 6815744 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7077888 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9175040 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_z.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 9437184 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_z.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 11534336 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11796480 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11812864 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11814912 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11831296 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11833344 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11849728 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11851776 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11868160 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11870208 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11886592 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11888640 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11905024 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11907072 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11923456 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11925504 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11941888 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11943936 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11960320 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11962368 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11978752 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11980800 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 11997184 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 11999232 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12015616 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12017664 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12034048 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12036096 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12052480 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12054528 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12070912 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12072960 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12089344 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12091392 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12107776 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12109824 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12126208 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12128256 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12144640 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12146688 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12163072 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12165120 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12181504 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12183552 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12199936 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12201984 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12218368 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12220416 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12236800 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12238848 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12255232 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12257280 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12273664 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12275712 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12292096 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12294144 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12310528 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12312576 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12328960 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12331008 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12347392 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12349440 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12365824 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12367872 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12384256 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12386304 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12402688 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12404736 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12421120 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_b.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12423168 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_b.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12439552 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_a.q_weight", | |
| "shape": [ | |
| 16, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16384, | |
| "byteOffset": 12441600 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.in_proj_a.q_scale", | |
| "shape": [ | |
| 16, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2048, | |
| "byteOffset": 12457984 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12460032 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12509184 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12558336 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12607488 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12656640 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12705792 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12754944 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12804096 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12853248 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12902400 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 12951552 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13000704 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13049856 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13099008 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13148160 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13197312 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13246464 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.conv1d_weight", | |
| "shape": [ | |
| 6144, | |
| 1, | |
| 4 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 49152, | |
| "byteOffset": 13295616 | |
| }, | |
| { | |
| "name": "model.layers.11.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13344768 | |
| }, | |
| { | |
| "name": "model.layers.11.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13348864 | |
| }, | |
| { | |
| "name": "model.layers.1.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13352960 | |
| }, | |
| { | |
| "name": "model.layers.10.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13357056 | |
| }, | |
| { | |
| "name": "model.layers.12.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13361152 | |
| }, | |
| { | |
| "name": "model.layers.16.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13365248 | |
| }, | |
| { | |
| "name": "model.layers.15.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13369344 | |
| }, | |
| { | |
| "name": "model.layers.18.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13373440 | |
| }, | |
| { | |
| "name": "model.layers.19.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13377536 | |
| }, | |
| { | |
| "name": "model.layers.0.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13381632 | |
| }, | |
| { | |
| "name": "model.layers.14.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13385728 | |
| }, | |
| { | |
| "name": "model.layers.1.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13389824 | |
| }, | |
| { | |
| "name": "model.layers.10.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13393920 | |
| }, | |
| { | |
| "name": "model.layers.14.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13398016 | |
| }, | |
| { | |
| "name": "model.layers.15.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13402112 | |
| }, | |
| { | |
| "name": "model.layers.19.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13406208 | |
| }, | |
| { | |
| "name": "model.layers.2.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13410304 | |
| }, | |
| { | |
| "name": "model.layers.20.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13414400 | |
| }, | |
| { | |
| "name": "model.layers.13.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13418496 | |
| }, | |
| { | |
| "name": "model.layers.13.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13422592 | |
| }, | |
| { | |
| "name": "model.layers.21.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13426688 | |
| }, | |
| { | |
| "name": "model.layers.22.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13430784 | |
| }, | |
| { | |
| "name": "model.layers.23.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13434880 | |
| }, | |
| { | |
| "name": "model.layers.12.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13438976 | |
| }, | |
| { | |
| "name": "model.layers.18.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13443072 | |
| }, | |
| { | |
| "name": "model.layers.16.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13447168 | |
| }, | |
| { | |
| "name": "model.layers.23.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13451264 | |
| }, | |
| { | |
| "name": "model.layers.3.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13455360 | |
| }, | |
| { | |
| "name": "model.layers.5.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13459456 | |
| }, | |
| { | |
| "name": "model.layers.20.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13463552 | |
| }, | |
| { | |
| "name": "model.layers.3.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13467648 | |
| }, | |
| { | |
| "name": "model.layers.21.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13471744 | |
| }, | |
| { | |
| "name": "model.layers.4.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13475840 | |
| }, | |
| { | |
| "name": "model.layers.22.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13479936 | |
| }, | |
| { | |
| "name": "model.layers.7.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13484032 | |
| }, | |
| { | |
| "name": "model.layers.2.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13488128 | |
| }, | |
| { | |
| "name": "model.layers.17.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13492224 | |
| }, | |
| { | |
| "name": "model.layers.17.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13496320 | |
| }, | |
| { | |
| "name": "model.layers.9.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13500416 | |
| }, | |
| { | |
| "name": "model.layers.8.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13504512 | |
| }, | |
| { | |
| "name": "model.layers.6.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13508608 | |
| }, | |
| { | |
| "name": "model.layers.7.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13512704 | |
| }, | |
| { | |
| "name": "model.layers.5.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13516800 | |
| }, | |
| { | |
| "name": "model.layers.4.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13520896 | |
| }, | |
| { | |
| "name": "model.layers.6.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13524992 | |
| }, | |
| { | |
| "name": "model.layers.8.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13529088 | |
| }, | |
| { | |
| "name": "model.layers.9.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13533184 | |
| }, | |
| { | |
| "name": "model.layers.0.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13537280 | |
| }, | |
| { | |
| "name": "model.norm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 13541376 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13545472 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13545984 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13546240 | |
| }, | |
| { | |
| "name": "model.layers.15.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13546752 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13547264 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13547520 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13547776 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13548032 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13548288 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13548544 | |
| }, | |
| { | |
| "name": "model.layers.19.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13549056 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13549568 | |
| }, | |
| { | |
| "name": "model.layers.11.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13549824 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13550336 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13550592 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13551104 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13551616 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13551872 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13552128 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13552640 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13552896 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13553408 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.k_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13553664 | |
| }, | |
| { | |
| "name": "model.layers.7.self_attn.q_norm.weight", | |
| "shape": [ | |
| 256 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 512, | |
| "byteOffset": 13554176 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13554688 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13554944 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13555200 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13555456 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13555712 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.norm.weight", | |
| "shape": [ | |
| 128 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 256, | |
| "byteOffset": 13555968 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556224 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556256 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556288 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556320 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556352 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556384 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556416 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556448 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556480 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556512 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556544 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556576 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556608 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556640 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556672 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556704 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556736 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.A_log", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556768 | |
| }, | |
| { | |
| "name": "model.layers.10.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556800 | |
| }, | |
| { | |
| "name": "model.layers.17.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556832 | |
| }, | |
| { | |
| "name": "model.layers.14.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556864 | |
| }, | |
| { | |
| "name": "model.layers.1.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556896 | |
| }, | |
| { | |
| "name": "model.layers.20.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556928 | |
| }, | |
| { | |
| "name": "model.layers.13.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556960 | |
| }, | |
| { | |
| "name": "model.layers.12.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13556992 | |
| }, | |
| { | |
| "name": "model.layers.18.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557024 | |
| }, | |
| { | |
| "name": "model.layers.16.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557056 | |
| }, | |
| { | |
| "name": "model.layers.21.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557088 | |
| }, | |
| { | |
| "name": "model.layers.4.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557120 | |
| }, | |
| { | |
| "name": "model.layers.22.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557152 | |
| }, | |
| { | |
| "name": "model.layers.2.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557184 | |
| }, | |
| { | |
| "name": "model.layers.5.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557216 | |
| }, | |
| { | |
| "name": "model.layers.6.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557248 | |
| }, | |
| { | |
| "name": "model.layers.8.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557280 | |
| }, | |
| { | |
| "name": "model.layers.9.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557312 | |
| }, | |
| { | |
| "name": "model.layers.0.linear_attn.dt_bias", | |
| "shape": [ | |
| 16 | |
| ], | |
| "dtype": "float32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32, | |
| "byteOffset": 13557344 | |
| } | |
| ], | |
| "md5sum": "9d55eec73276417695dfab3e8d21c560" | |
| } | |
| ] | |
| } |