diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..eda691a7591e154560985941ac9309c45aff1305 --- /dev/null +++ b/README.md @@ -0,0 +1,52 @@ +--- +library_name: mlc-llm +base_model: nvidia/OpenCodeReasoning-Nemotron-1.1-32B +tags: +- mlc-llm +- web-llm +--- + +# OpenCodeReasoning-Nemotron-1.1-32B-q3f16_1-MLC + +This is the [OpenCodeReasoning-Nemotron-1.1-32B](https://huggingface.co/nvidia/OpenCodeReasoning-Nemotron-1.1-32B) model in MLC format `q3f16_1`. +The model can be used with [MLC-LLM](https://github.com/mlc-ai/mlc-llm) and [WebLLM](https://github.com/mlc-ai/web-llm). + +## Example Usage + +Before using the examples, please follow the [installation guide](https://llm.mlc.ai/docs/install/mlc_llm.html#install-mlc-packages). + +### Chat CLI + +```bash +mlc_llm chat HF://JackBinary/OpenCodeReasoning-Nemotron-1.1-32B-q3f16_1-MLC +```` + +### REST Server + +```bash +mlc_llm serve HF://JackBinary/OpenCodeReasoning-Nemotron-1.1-32B-q3f16_1-MLC +``` + +### Python API + +```python +from mlc_llm import MLCEngine + +model = "HF://JackBinary/OpenCodeReasoning-Nemotron-1.1-32B-q3f16_1-MLC" +engine = MLCEngine(model) + +for response in engine.chat.completions.create( + messages=[{"role": "user", "content": "What is the meaning of life?"}], + model=model, + stream=True, +): + for choice in response.choices: + print(choice.delta.content, end="", flush=True) +print("\n") + +engine.terminate() +``` + +## Documentation + +For more on MLC LLM, visit the [documentation](https://llm.mlc.ai/docs/) and [GitHub repo](https://github.com/mlc-ai/mlc-llm). diff --git a/ndarray-cache.json b/ndarray-cache.json new file mode 100644 index 0000000000000000000000000000000000000000..dc1e66891631d4327a12228bbfe465c87e1715ac --- /dev/null +++ b/ndarray-cache.json @@ -0,0 +1,9647 @@ +{ + "metadata": { + "ParamSize": 709, + "ParamBytes": 14750197760.0, + "BitsPerParam": 3.135010764942777 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 311427072, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 152064, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 311427072, + "byteOffset": 0 + } + ], + "md5sum": "e7f60bb591807cc61be9838fb00d843b" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 38928384, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 152064, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 38928384, + "byteOffset": 0 + } + ], + "md5sum": "15c3569e2847d9637ef66ea0333f35e8" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.58.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "04dfd1784edad186f584c8a563dc9aaa" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.58.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "2edaf63324cef43cf357eed1aaa6f0a0" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.59.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "916ba9fc52ca1693776f20ef0e8107e4" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.59.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "49d2c24f0393cdde4e7fd5b37d6b10fa" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 28358656, + "records": [ + { + "name": "model.layers.58.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 0 + }, + { + "name": "model.layers.58.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 10240 + }, + { + "name": "model.layers.58.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7096320 + }, + { + "name": "model.layers.58.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21252096 + }, + { + "name": "model.layers.59.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21262336 + }, + { + "name": "model.layers.59.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 21272576 + } + ], + "md5sum": "5fdcba86134220b4213283fdd631cefd" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.59.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.59.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.59.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.59.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.59.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "f83b5a15f7ffd3c34c57f0e8c2cf7bdd" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.60.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "5ba66388321850591ba3d5b9206f4fa5" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.60.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "08bbf323725a2ba4ff73b7add24344d8" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.59.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.59.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.60.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.60.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.60.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.60.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.60.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "8f4de071b97de34332a1623f2cc9b7e9" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.61.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "f9a2a21efe41670a25d84bca4d0dc83e" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.60.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.60.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.60.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.60.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.61.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "742df16527aaa88d3a20bd053f4e322a" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.61.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "d3a38e6beeaad58ecce016ec3ad73cb9" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.61.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.61.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.61.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.61.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "7810941a9c826b99220cd167441ee69b" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.62.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "58680edc27a77f5c066e92302c1c735d" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.61.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.61.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.61.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.61.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.62.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "54fcf7ccc60f565f39c2477ae6a047d9" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.62.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "4ece8afc4bfbd60bdabed4ae91945888" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.62.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.62.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.62.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.62.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "39e0f62da5cb9737f5e03e609f08587f" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.63.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "994ede9e229b079c2c824af6b05acf86" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.62.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.62.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.62.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.62.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.63.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "64b9dbc0922cbb9e1a97fe99228e02f0" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.63.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "d2d91924d3c61b0f1e226ed46fa269c4" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.63.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.63.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.63.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.63.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "2b5aaed59a87bfbd2c85705ec28de505" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 311427072, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 152064, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 311427072, + "byteOffset": 0 + } + ], + "md5sum": "5dab721aa5048f2bf4b5963875710b9e" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 38928384, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 152064, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 38928384, + "byteOffset": 0 + } + ], + "md5sum": "c3541a5ad806b7a39ae75e7d385b3c63" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "190c58971fd251822385f91993c3e64d" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 28332032, + "records": [ + { + "name": "model.layers.63.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.63.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.63.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.63.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.norm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28321792 + } + ], + "md5sum": "095ddfe3646e9a4c1d68b5f80137478b" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "a60716f7ba3d8a7c7b5d8c19106d5af7" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "6e4ba1ebe55363de674f9a7cd3fd59fb" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "ff95205bbae3ca185498095e4f56fd0a" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.0.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "2a57f63a7d0648966d12bece029589ec" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "28f3fc6cb27afa155c73b24cc253d902" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "867fe1493bfc4149a3df8656b6fcee1e" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "26ccdea2d9d0b10071095d7499e38923" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.1.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "3b5d5229f24792e1f9e980b9dd0e6534" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "70756f039f6b75ea42379fdbeadb055c" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "a3d51aecd47f771d98764caeb1b809ee" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "131ba17f6afcc1fcac43823965deb195" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.2.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "2d9fd91c85e2add6d3416d35c581a9b6" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "64a084844f298ff7e41f90defc05809a" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "2db870fd861b86f19da8f06ca9eaaaa0" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "2ce7ba71b9c5f236ccbefe880643531f" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.3.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "b9af5684bfd8d329d791757fa233e0f3" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "7c580695c8fa500e778d70cea6c6b328" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "5aeab0c4f921ba02496e0bf88154055e" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "0643dc15875dc485e1dcb124aaf9b08c" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.4.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "b83615233cd79d207d3d91fb36df9e15" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "5ad623a9cdd3f46046272568388c38ca" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "7215e4635f29d7f512bbe12b0ffd7b85" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "45bd43c707f6be1adbf0c3de251ba765" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.5.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "9f571b29180b2badb96976c7872bcc78" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "6800dc3ba713b099c432cb7c0883e43e" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "485cc8c937491fc8073191ba516f4da3" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "8c98c0f051716038b1a0e86343e126be" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.6.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "a310e15f9878094bbdf929c8994a41d8" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "502cd97a8f4c78107a0e344e5a41aa48" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "7de30b877bfcaa7372fe870cea0cd334" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "9fd589cba2dda8115fe2f70c061029da" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.7.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "f352f8837dd35164a6c4d0a3cf98bf62" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 30685184, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14155776 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14170112 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28850176 + } + ], + "md5sum": "986ac0ca913d0c415eb6bf681ac9728b" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "25bc53aeb2c260033152b5830b92a0bc" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "4656ef0e8b8e8890551154eb14084a73" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "c450362eeb9c36920e0af38e053b9077" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "a7c12b3b2606959b835da0f485a2e4be" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.10.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "795ccaf1b2fba6913b3d40df98a6ee2b" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "a53f32a97c297f3d4ff3120e2dbfbbf3" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "5dfa7eec4da226b931eb3581bf85bc0a" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "e174d3cc8d07a90b62f4034982640531" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.11.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "0b8c5dfc8eb22952ee6350dc586a471b" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "63d187c9c041e4a84b25f2f6aebc7443" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "44abe629bea018c8dba18d54fb3db6a8" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "6b7713531640e4aa4d22660dfd4b1501" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.12.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "9baa2996bde1948a1226e46f0fcdc23b" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "c756deef51ffca72cec66aed4625c590" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "b8f83a00474592dd05e0f5c5440670b2" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "c1ccd289dc6c88cbb7f34a4b9869aa34" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.13.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "5beba696424ede4ce80a370507cc8a13" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "011837cf1fcf970690d53a980c69a72c" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "aca2d278418fc93b501afc7423d6febb" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "214db1f51b58f167229e7c70e5d7e744" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.14.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "eb6ca173644aa77237e703a51e67fd6e" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "458c4d220e938e36ef69a6cc37e3dcf8" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "f36975f02135b874c08e6af05ed2fda2" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "816ee359c0ded5db736a08837a0578a3" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.15.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "1a1dc6cc87d8ebaa15e25119050e1e05" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "c60fce92eef821e0faad9501f9e92142" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "4659819bae51935dc68f6436db55bf8a" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "9fc910b193c90997dbc40c3d7f387424" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.16.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "4e86151ed802fb4589b45f51ace8f292" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "5163b718f270cf34b900c86b75e6837b" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "be97392ef63bed2e30723e92ea25e685" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "8a5f1af381e1b91bcab50b747ac27deb" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.17.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "f3ac9c05258aac6e54355443988eac82" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 30685184, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14155776 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14170112 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28850176 + } + ], + "md5sum": "07cd9256465a891e5e872820431d7b72" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "0e979c304d4b91425dc8cfb9e010f753" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "b12cda1b92b9fd38d36fcc5efab4a5a3" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "0869d16a90349fc969563590227dcefa" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 25999360, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18892800 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18903040 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 18913280 + } + ], + "md5sum": "99386a29e76e37a504319a4dc8638753" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "f889ce4bee3bd35cead93c6c9990168d" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "ebd22148c1f82bfbea7b6ead25a56a5b" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "37e11a17404f9aa682a8cd5d86e88ea0" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "b6783a8230bd38c6c6952790e224beae" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 25999360, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18892800 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18903040 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 18913280 + } + ], + "md5sum": "c9a232410206c7d3bfaa8061ca9292e2" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "b720c8525a464ec0cf6663cdd96981d3" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "db9d815a7e1ed6b467261f3cd6858dfa" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "a012b35c728852707caa96b538e8a7e6" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "f640cdd64f6fafa4b72e1a2dfccec2ba" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "148c58818d282d658a14b6171e44f348" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.20.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "3f2664774e71e392759f0f4edd3755c2" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "de9c2f1e153e14dabbd2c9c8788d7c26" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "aa29d6821b1a1d5bbb842a1fe224ddd0" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "e80848562bd1022fe16732ed57e9ac74" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.21.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "3a612824e6c13c83e23886b53570eeda" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "75e6cfb82769387fc465110a08f191fd" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "f2fca26b8d9eba2635b5c5516b405f66" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "30252f7f548aac7fa8bf3417ecca922f" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.22.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "6033268ac588b3959d862b19ff24598b" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "98ae1821313bbe959e28149e1cea0571" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "d271785b4968ac8914d94f638ebed41a" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "05c4e410c8b28931763c3de07086fd88" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.23.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "be326cd4c6aaa015f10c44f1712f06af" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "43c5c300fc5217219d9f1d2efaf6ad21" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "b8e1135a7e1a210a0d608676f8a4e9c9" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "6a6011c3bd034bc77e8b567ffaac8b67" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.24.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "5e88bc6112da02ca1e3bbefc3d2b7633" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "b381a46612bb53196149f12442c1aa80" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "a5177c001df684e89386d430f3e637d6" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "a6bc048c3e483a50c3d4f2c6fbb5d0eb" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.25.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "418d71b31e336b511ce6c49f9e1a2f3e" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "7c2d76663e9163a4e47d0a34657b6a36" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "326c654a3974a4cf4168d08edb394998" + }, + { + "dataPath": "params_shard_131.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "5c85854e61e56992e72821a23734ad64" + }, + { + "dataPath": "params_shard_132.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.26.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "8860da0f41724acee9358c1697a67072" + }, + { + "dataPath": "params_shard_133.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "b760843a0b70af0c1e9f2affd6db0c17" + }, + { + "dataPath": "params_shard_134.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "0272e39929b45696c14021c10fc776c8" + }, + { + "dataPath": "params_shard_135.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "d78f4d430f1f41e5ec164ec3facb1b5a" + }, + { + "dataPath": "params_shard_136.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.27.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "68cedd8b0faacf49f6e1490561a2cb4c" + }, + { + "dataPath": "params_shard_137.bin", + "format": "raw-shard", + "nbytes": 30685184, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14155776 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14170112 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28850176 + } + ], + "md5sum": "7c4b3cf6f4ae09f29269e63e57959b6f" + }, + { + "dataPath": "params_shard_138.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "fd5cab861088f655a168b56a22e85ea1" + }, + { + "dataPath": "params_shard_139.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "02aad305ced41491b0f638782d5a0861" + }, + { + "dataPath": "params_shard_140.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "14c862f17d8f57f88ebb9305d0376520" + }, + { + "dataPath": "params_shard_141.bin", + "format": "raw-shard", + "nbytes": 25999360, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18892800 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18903040 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 18913280 + } + ], + "md5sum": "0ae254b335dd1a113ecf717cd85483ac" + }, + { + "dataPath": "params_shard_142.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.29.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "0006861738ada3857fa1ac29bbd66d45" + }, + { + "dataPath": "params_shard_143.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "5b3e8f60f214fe0e0e796ecf98f61693" + }, + { + "dataPath": "params_shard_144.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "2d6bcfa418ce81eb8427a3dfe6193e1e" + }, + { + "dataPath": "params_shard_145.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.30.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "c08ec3651aff4847c453c43dc273d6ef" + }, + { + "dataPath": "params_shard_146.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "f36a2b6f305a7a668ff9aebc17ed484b" + }, + { + "dataPath": "params_shard_147.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.30.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "0039c4c877fa06db4e99b6443cd3fcce" + }, + { + "dataPath": "params_shard_148.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "3dda87e7a020db79a75a24c150ae5ec9" + }, + { + "dataPath": "params_shard_149.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.31.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "ad1da73c41e67246a315fa083a4fa92d" + }, + { + "dataPath": "params_shard_150.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "f21ca11049c57b9252f565f798da516d" + }, + { + "dataPath": "params_shard_151.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.31.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.31.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "c84f237c0fe457c36b2f1c4b98ae0395" + }, + { + "dataPath": "params_shard_152.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "0d30033a25b3a1593b138fedf73eb227" + }, + { + "dataPath": "params_shard_153.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.32.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.32.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "78461551dce58185dd9d8c0fe2e4ea7f" + }, + { + "dataPath": "params_shard_154.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "4805ac2db1c1fe7a0bf3ff9004a636b2" + }, + { + "dataPath": "params_shard_155.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.32.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.32.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "0a5e6760571ad1142b2a6a337f3fb00f" + }, + { + "dataPath": "params_shard_156.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "d6adefa75fe46acb12cfac85c97c5f45" + }, + { + "dataPath": "params_shard_157.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.33.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.33.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "ba2124bf820012de9f62fc2f60a2cf1a" + }, + { + "dataPath": "params_shard_158.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "bb6a59961b6eecb19c947531903267d6" + }, + { + "dataPath": "params_shard_159.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.33.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.33.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "15f455c3ca208c45f9fea8faca5dc191" + }, + { + "dataPath": "params_shard_160.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "6c82f180dbfa7cb5b762159677fecfba" + }, + { + "dataPath": "params_shard_161.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.34.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.34.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "ee2ede6349a028e9568b239cdac8c79a" + }, + { + "dataPath": "params_shard_162.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "bbd1945cf637dd834c18c2d2b72ee7f6" + }, + { + "dataPath": "params_shard_163.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.34.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.34.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "1074ce7232df74fdd97d4b85a080bfc4" + }, + { + "dataPath": "params_shard_164.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "7f83acd5dc182598bea637901c5671fc" + }, + { + "dataPath": "params_shard_165.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.35.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.35.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "0e55adff1269484526989e3193ff1109" + }, + { + "dataPath": "params_shard_166.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "b883e8394b383dade511ac7ee1618457" + }, + { + "dataPath": "params_shard_167.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.35.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.35.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.36.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "62556531d21b2da347e2585d12f322e9" + }, + { + "dataPath": "params_shard_168.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.36.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "7396eb383dfc8bd6a46a678805ca0de5" + }, + { + "dataPath": "params_shard_169.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.36.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.36.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.36.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "748b37448a6306f1f0df756803146b98" + }, + { + "dataPath": "params_shard_170.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "3c8c1fa4d8de5dbe4ebd1bb497233b07" + }, + { + "dataPath": "params_shard_171.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.36.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.36.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.36.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.36.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.37.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "8da899702a0b9bd406c4a6b76d12a4e0" + }, + { + "dataPath": "params_shard_172.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.37.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "4ee9a8757c33f79863460e8dbdcb7425" + }, + { + "dataPath": "params_shard_173.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.37.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.37.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.37.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "88550fae4a0ed44218fb0fa7cd2aaded" + }, + { + "dataPath": "params_shard_174.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "2984257405c02e994f469082bbc307c9" + }, + { + "dataPath": "params_shard_175.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.37.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.37.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.37.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.37.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "ac1294e4de5d19076de5733c17057178" + }, + { + "dataPath": "params_shard_176.bin", + "format": "raw-shard", + "nbytes": 30685184, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.38.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14155776 + }, + { + "name": "model.layers.38.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14170112 + }, + { + "name": "model.layers.38.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28850176 + } + ], + "md5sum": "535288d8f76b748776b536f21fb5fb92" + }, + { + "dataPath": "params_shard_177.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.38.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "75d6d4d55dcf96a52328011e15d4fc8f" + }, + { + "dataPath": "params_shard_178.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.39.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "98cb24bc5ba16911f51792931066bb16" + }, + { + "dataPath": "params_shard_179.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "57dd47fd0621da07fab3bdaf625821dd" + }, + { + "dataPath": "params_shard_180.bin", + "format": "raw-shard", + "nbytes": 25999360, + "records": [ + { + "name": "model.layers.38.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.38.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.38.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.38.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.38.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18892800 + }, + { + "name": "model.layers.39.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18903040 + }, + { + "name": "model.layers.39.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 18913280 + } + ], + "md5sum": "b4b67cbac48ce7d57ab65da7c963cb57" + }, + { + "dataPath": "params_shard_181.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.39.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.39.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.39.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.39.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "f0a6c6136d414ce661b965884c901962" + }, + { + "dataPath": "params_shard_182.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.40.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "bbc0f6a6840f73020a199b1f43711992" + }, + { + "dataPath": "params_shard_183.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.40.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "d2e1a2d39e730b4a7610609d180bc4b1" + }, + { + "dataPath": "params_shard_184.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.39.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.39.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.40.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.40.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.40.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.40.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.40.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "8eaf357b60d40347bb29efdbffb7437f" + }, + { + "dataPath": "params_shard_185.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "deafc1ad1ba672965f71f78c2eff3c01" + }, + { + "dataPath": "params_shard_186.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.40.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.40.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.40.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.40.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.41.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "abd29d161450b73ef732237cdebfb040" + }, + { + "dataPath": "params_shard_187.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.41.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "fc1459e0471d6adff0b83be2881552fa" + }, + { + "dataPath": "params_shard_188.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.41.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.41.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.41.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "b66c36a61a7d3c4ce069e7fc293bf7b2" + }, + { + "dataPath": "params_shard_189.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "c70ea569ec747f06b2af7dfdbff5d0a9" + }, + { + "dataPath": "params_shard_190.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.41.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.41.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.41.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.41.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.42.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "780491b73d30a325c80749c0b8d56617" + }, + { + "dataPath": "params_shard_191.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.42.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "f3e955fe1ddca8095d0a0642e7b39a98" + }, + { + "dataPath": "params_shard_192.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.42.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.42.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.42.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "78c8c6a04923652b2ffaf97bedd2deaa" + }, + { + "dataPath": "params_shard_193.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "e4d8d00b73f1bd497c2415cf7c07f0d5" + }, + { + "dataPath": "params_shard_194.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.42.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.42.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.42.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.42.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.43.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "6510b9b028488f814ff6b5b0edfd3261" + }, + { + "dataPath": "params_shard_195.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.43.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "841341c68e444c153fedfae148ca0624" + }, + { + "dataPath": "params_shard_196.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.43.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.43.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.43.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "9f50ab51bfbfd53b812e141cd4f67d58" + }, + { + "dataPath": "params_shard_197.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "75779ba08f10e1b5979a50111a82e6fa" + }, + { + "dataPath": "params_shard_198.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.43.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.43.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.43.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.43.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.44.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "2b4215d85aedcc7b1d09371fb7a1d213" + }, + { + "dataPath": "params_shard_199.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.44.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "8e0daaf2f9e853cdf5d3ddf627c6b452" + }, + { + "dataPath": "params_shard_200.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.44.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.44.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.44.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "ee6d1badf447aa1913aae482fa83b298" + }, + { + "dataPath": "params_shard_201.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "acd937269917693393dd95ab2f34002d" + }, + { + "dataPath": "params_shard_202.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.44.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.44.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.44.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.44.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.45.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "66b24720b92fc8c5bceab22fe1155b26" + }, + { + "dataPath": "params_shard_203.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.45.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "1be262b07f2ca460306554f8799ac7a3" + }, + { + "dataPath": "params_shard_204.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.45.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.45.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.45.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "7d7453010bf7fd912494a9b6245a161b" + }, + { + "dataPath": "params_shard_205.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "fec8fc8382a86d3aea25a39c37761061" + }, + { + "dataPath": "params_shard_206.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.45.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.45.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.45.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.45.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.46.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "7fa978512ba8624ef77abd21dd05796c" + }, + { + "dataPath": "params_shard_207.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.46.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "43b9891c203436d15c8b49f01fe296a3" + }, + { + "dataPath": "params_shard_208.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.46.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.46.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.46.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "e814e09423feb15a48ea4232dd3148f0" + }, + { + "dataPath": "params_shard_209.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "2ad741c39183c0d6b98415a7929fb378" + }, + { + "dataPath": "params_shard_210.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.46.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.46.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.46.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.46.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.47.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "8519c594944fb86e76f2341d1841873d" + }, + { + "dataPath": "params_shard_211.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.47.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "388f3289b24164cae63b5f0359558b95" + }, + { + "dataPath": "params_shard_212.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.47.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.47.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.47.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "20d25404b25702835c4cf2520592a730" + }, + { + "dataPath": "params_shard_213.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.48.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "f4eaadda3030a31e87a4ba259564e954" + }, + { + "dataPath": "params_shard_214.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.47.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.47.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.47.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.47.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "5e7fe29adbfa6520bc01e7ca407d0f91" + }, + { + "dataPath": "params_shard_215.bin", + "format": "raw-shard", + "nbytes": 30685184, + "records": [ + { + "name": "model.layers.48.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.48.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14155776 + }, + { + "name": "model.layers.48.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14170112 + }, + { + "name": "model.layers.48.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28850176 + } + ], + "md5sum": "c8dec5ca5b625988d5c7b05915bfd618" + }, + { + "dataPath": "params_shard_216.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.48.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "744b1272e6e7b1b209b1d39c2cd67a6f" + }, + { + "dataPath": "params_shard_217.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.49.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "4674455c8703f61f5b04ea9bac225d75" + }, + { + "dataPath": "params_shard_218.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.49.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "89de9c17460288ad4bcc9fd5c0e5c0f9" + }, + { + "dataPath": "params_shard_219.bin", + "format": "raw-shard", + "nbytes": 25999360, + "records": [ + { + "name": "model.layers.48.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.48.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.48.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.48.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.48.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18892800 + }, + { + "name": "model.layers.49.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 18903040 + }, + { + "name": "model.layers.49.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 18913280 + } + ], + "md5sum": "3fdd7d9275856b51894322d6aeb5e6c6" + }, + { + "dataPath": "params_shard_220.bin", + "format": "raw-shard", + "nbytes": 30695424, + "records": [ + { + "name": "model.layers.49.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 0 + }, + { + "name": "model.layers.49.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 14155776 + }, + { + "name": "model.layers.49.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 14166016 + }, + { + "name": "model.layers.49.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 14180352 + }, + { + "name": "model.layers.49.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 28860416 + } + ], + "md5sum": "9b96110c1c8f7ea605aac5fe8f9a9bee" + }, + { + "dataPath": "params_shard_221.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.50.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "c700344400934fa4d2971ff8fc8ce080" + }, + { + "dataPath": "params_shard_222.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.50.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "97c1dd77dfab2713ac74a022c406fea5" + }, + { + "dataPath": "params_shard_223.bin", + "format": "raw-shard", + "nbytes": 33073152, + "records": [ + { + "name": "model.layers.49.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 0 + }, + { + "name": "model.layers.49.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 10485760 + }, + { + "name": "model.layers.50.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 11796480 + }, + { + "name": "model.layers.50.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 11806720 + }, + { + "name": "model.layers.50.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 18892800 + }, + { + "name": "model.layers.50.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 33048576 + }, + { + "name": "model.layers.50.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 33058816 + } + ], + "md5sum": "d52fe0303eed65987d21bf09e7aa2d87" + }, + { + "dataPath": "params_shard_224.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.51.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "e856516fe522f8714fb3ea86c54caf4c" + }, + { + "dataPath": "params_shard_225.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.50.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.50.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.50.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.50.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.51.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "4d110ce045cc06a4e7c61b4e53abd670" + }, + { + "dataPath": "params_shard_226.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.51.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "59a7174f92a5ec5015977405ef9ad398" + }, + { + "dataPath": "params_shard_227.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.51.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.51.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.51.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.51.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "4acaf02344f3272dc9bf0330bf6c64c0" + }, + { + "dataPath": "params_shard_228.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.52.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "57696dc7bbcf5f51f35d7a919077552f" + }, + { + "dataPath": "params_shard_229.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.51.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.51.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.51.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.51.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.52.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "4b8934d595c196f84a9ac9cd4ce21ae4" + }, + { + "dataPath": "params_shard_230.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.52.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "90a6962097d509c88247b49631140153" + }, + { + "dataPath": "params_shard_231.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.52.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.52.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.52.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.52.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "8e8e396fc338c67535bfbb51b6109c9b" + }, + { + "dataPath": "params_shard_232.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.53.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "61ced095b82b2569f98d33ad7ff8543a" + }, + { + "dataPath": "params_shard_233.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.52.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.52.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.52.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.52.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.53.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "eaa12f7236e0da024a40275aae1a1c07" + }, + { + "dataPath": "params_shard_234.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.53.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "fc66de9c5eb71a0ac979ce010d2da57f" + }, + { + "dataPath": "params_shard_235.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.53.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.53.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.53.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.53.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "028c7d5dee1062f33c1099f025370503" + }, + { + "dataPath": "params_shard_236.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.54.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "d65ea17a8c0b2095cf1c9a7a71e9f87d" + }, + { + "dataPath": "params_shard_237.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.53.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.53.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.53.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.53.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.54.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "d878976aebf9f4e74295e26ca5200882" + }, + { + "dataPath": "params_shard_238.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.54.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "142936d608ce50fa61069c9c0e56b3f1" + }, + { + "dataPath": "params_shard_239.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.54.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.54.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.54.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.54.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "6081dd6b9c21213860b2227c121b0a34" + }, + { + "dataPath": "params_shard_240.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.55.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "db27e558bcb3001aa3d5404de4cefa7e" + }, + { + "dataPath": "params_shard_241.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.54.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.54.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.54.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.54.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.55.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "74f9b218b6645f5918652777e3275b26" + }, + { + "dataPath": "params_shard_242.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.55.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "be499c853a0a51a3d1d96a286aceeb47" + }, + { + "dataPath": "params_shard_243.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.55.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.55.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.55.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.55.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "3d32aa54c37c717c931b9320023c603e" + }, + { + "dataPath": "params_shard_244.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.56.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "0784f5f5257be13d509a30663ca66a3f" + }, + { + "dataPath": "params_shard_245.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.55.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.55.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.55.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.55.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.56.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "73d70295f0bf3927fec50a0cf0329819" + }, + { + "dataPath": "params_shard_246.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.56.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "1617f1432405ae88a799c20faf00dbde" + }, + { + "dataPath": "params_shard_247.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.56.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.56.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.56.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.56.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "76ffac9e5b7b68ab0cd6d44360d1a549" + }, + { + "dataPath": "params_shard_248.bin", + "format": "raw-shard", + "nbytes": 56688640, + "records": [ + { + "name": "model.layers.57.mlp.down_proj.q_weight", + "shape": [ + 5120, + 2768 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 56688640, + "byteOffset": 0 + } + ], + "md5sum": "d8052882fb88367d0503a762619896b5" + }, + { + "dataPath": "params_shard_249.bin", + "format": "raw-shard", + "nbytes": 28321792, + "records": [ + { + "name": "model.layers.56.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.56.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.56.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.56.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.57.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 28311552 + } + ], + "md5sum": "dee0d1ba07aa44a163ed3709181ae6a4" + }, + { + "dataPath": "params_shard_250.bin", + "format": "raw-shard", + "nbytes": 113246208, + "records": [ + { + "name": "model.layers.57.mlp.gate_up_proj.q_weight", + "shape": [ + 55296, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 113246208, + "byteOffset": 0 + } + ], + "md5sum": "1791cf3464c577242d31e0d7ad47dd77" + }, + { + "dataPath": "params_shard_251.bin", + "format": "raw-shard", + "nbytes": 21266432, + "records": [ + { + "name": "model.layers.57.mlp.down_proj.q_scale", + "shape": [ + 5120, + 692 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7086080, + "byteOffset": 0 + }, + { + "name": "model.layers.57.mlp.gate_up_proj.q_scale", + "shape": [ + 55296, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14155776, + "byteOffset": 7086080 + }, + { + "name": "model.layers.57.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 21241856 + }, + { + "name": "model.layers.57.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 21252096 + } + ], + "md5sum": "2c3d56f1aec91ab8bec81cdc5f3d2d4e" + }, + { + "dataPath": "params_shard_252.bin", + "format": "raw-shard", + "nbytes": 28325888, + "records": [ + { + "name": "model.layers.57.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.57.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.57.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.57.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + }, + { + "name": "model.layers.58.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 28311552 + } + ], + "md5sum": "a4810cea78624c1c5dcc995f588e3385" + }, + { + "dataPath": "params_shard_253.bin", + "format": "raw-shard", + "nbytes": 28311552, + "records": [ + { + "name": "model.layers.58.self_attn.c_attn.q_weight", + "shape": [ + 7168, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.58.self_attn.c_attn.q_scale", + "shape": [ + 7168, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 14680064 + }, + { + "name": "model.layers.58.self_attn.o_proj.q_weight", + "shape": [ + 5120, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 10485760, + "byteOffset": 16515072 + }, + { + "name": "model.layers.58.self_attn.o_proj.q_scale", + "shape": [ + 5120, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1310720, + "byteOffset": 27000832 + } + ], + "md5sum": "7d5b9405918d6a55bbb8ee7b03154185" + } + ] +} \ No newline at end of file diff --git a/params_shard_0.bin b/params_shard_0.bin new file mode 100644 index 0000000000000000000000000000000000000000..c45c4472790c0572c74de35106c7da67f950cf86 --- /dev/null +++ b/params_shard_0.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:efafd329faedd0467243cdca18e2162cf056807c5180dad741f9a838e4ca3afd +size 311427072 diff --git a/params_shard_102.bin b/params_shard_102.bin new file mode 100644 index 0000000000000000000000000000000000000000..c5237a208ffed0816d2b59042f86a1fba784facf --- /dev/null +++ b/params_shard_102.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fba46b116c8bd7be889ba006ea984cc25cebb3ffdf3cd3ab3d2903de0014821a +size 25999360 diff --git a/params_shard_103.bin b/params_shard_103.bin new file mode 100644 index 0000000000000000000000000000000000000000..1636a9715808c87f5e5e0896b8a51c9bdcf857ff --- /dev/null +++ b/params_shard_103.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81088a7b97d0856f35ee892cfc0244a791c4ca147d767871a6f81d0e59c0c7a5 +size 30695424 diff --git a/params_shard_107.bin b/params_shard_107.bin new file mode 100644 index 0000000000000000000000000000000000000000..03b0854cbe601b7d67932de1f9c76abb41476240 --- /dev/null +++ b/params_shard_107.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:605564d56d68918261763754deb97e4e2eeb768514d394d0c78680983143e7ee +size 56688640 diff --git a/params_shard_108.bin b/params_shard_108.bin new file mode 100644 index 0000000000000000000000000000000000000000..5f400b829001fb19935cc9d891e2caca4ccdb74a --- /dev/null +++ b/params_shard_108.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e061119b4b72c867d1838ef4976fc349f5a8e37ce4ca74aa50d70be95e40b12 +size 28321792 diff --git a/params_shard_111.bin b/params_shard_111.bin new file mode 100644 index 0000000000000000000000000000000000000000..180bbc664bd433e2705c1afc4aa01a88c2d9536c --- /dev/null +++ b/params_shard_111.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a1d1eb8e550b7e01e659b1410eb934a81ddf54c42df27df5d54924462c85920 +size 56688640 diff --git a/params_shard_114.bin b/params_shard_114.bin new file mode 100644 index 0000000000000000000000000000000000000000..acdc8b4ceaffdfc0482fd3e35b7ce67ff5894a23 --- /dev/null +++ b/params_shard_114.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1153a96e452331bb9e6361030c94d24e6df63da4658b9c60d56620be92c53ae8 +size 21266432 diff --git a/params_shard_115.bin b/params_shard_115.bin new file mode 100644 index 0000000000000000000000000000000000000000..ae2b3b5659270a0b6db2bdc24d3edf7292a18d32 --- /dev/null +++ b/params_shard_115.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9a5c7be9b71b6d06c5cab31c644702edf50135729f8bf71539ad4c9ded6e7ab +size 56688640 diff --git a/params_shard_117.bin b/params_shard_117.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d0bc15ef92b0609473fd5521bbdb4163264a2b5 --- /dev/null +++ b/params_shard_117.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6836b404edea90e987ab42e1bd16df9747275f067aa045310db058460caee8b1 +size 113246208 diff --git a/params_shard_121.bin b/params_shard_121.bin new file mode 100644 index 0000000000000000000000000000000000000000..6789335183b4ad2b5fb5973bb65577c72418ed4c --- /dev/null +++ b/params_shard_121.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:324f1eba8e298def9e4f553e90a7866bf54d79aacddb1186ae35121f09f0fd34 +size 113246208 diff --git a/params_shard_122.bin b/params_shard_122.bin new file mode 100644 index 0000000000000000000000000000000000000000..2dcf9a4e118a5a468fbc8747109b4df09430815f --- /dev/null +++ b/params_shard_122.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eea1fd3a3c49b0379a8939f6f242752036679d3e770c8361758d3750d23b8935 +size 21266432 diff --git a/params_shard_127.bin b/params_shard_127.bin new file mode 100644 index 0000000000000000000000000000000000000000..f6510ad314318cbb174ce7c4bc7cd9ee2081c916 --- /dev/null +++ b/params_shard_127.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88743e6d6c318ac5e40d6aed91ad7f56d48425598fb066b9d2e58ce7f46de4a0 +size 56688640 diff --git a/params_shard_129.bin b/params_shard_129.bin new file mode 100644 index 0000000000000000000000000000000000000000..dfaefa356d30168fa11262a8641db72ca6aae4f2 --- /dev/null +++ b/params_shard_129.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a93b58e6429a57c9a519404d95c33fa80b765638f6620d019b0c192792926acc +size 113246208 diff --git a/params_shard_131.bin b/params_shard_131.bin new file mode 100644 index 0000000000000000000000000000000000000000..7ae5f9306653918f136a02e18903be2ab9c31c31 --- /dev/null +++ b/params_shard_131.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e68ea2b645a8b86ab4007daefa8ac1a3d79183c9fca0b4548ba280689e202f7c +size 56688640 diff --git a/params_shard_133.bin b/params_shard_133.bin new file mode 100644 index 0000000000000000000000000000000000000000..aadfbffb7fdd129b135f7be975431b8838db8e84 --- /dev/null +++ b/params_shard_133.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10d0d706d9b004ed23537af52c1e551d240d92d402175a9e5822e81a11287ab5 +size 113246208 diff --git a/params_shard_135.bin b/params_shard_135.bin new file mode 100644 index 0000000000000000000000000000000000000000..86d82d504bf7cd106d6db549cd1ba2e412a439ab --- /dev/null +++ b/params_shard_135.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:436e6dee246b8b94a55ed6e0935de481450c30dbcf088349c9f1d7b7680a952b +size 113246208 diff --git a/params_shard_139.bin b/params_shard_139.bin new file mode 100644 index 0000000000000000000000000000000000000000..25d47a7730a177f31521d99da6a34591b20a057e --- /dev/null +++ b/params_shard_139.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ad46e94d4ddc05ec5779b49ab9600efccb4382facd401301e1ed4ead18fd434 +size 56688640 diff --git a/params_shard_14.bin b/params_shard_14.bin new file mode 100644 index 0000000000000000000000000000000000000000..e1b7870d97fecbbb54d916847d40a0cf2db8a28d --- /dev/null +++ b/params_shard_14.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6409db049bfb9905b24982b8e1d972bfdbd7db0ca57926f7e886c3202a7d8c45 +size 21266432 diff --git a/params_shard_140.bin b/params_shard_140.bin new file mode 100644 index 0000000000000000000000000000000000000000..2b211ff6025954f08df63ba0bae63a90a2cbc2a2 --- /dev/null +++ b/params_shard_140.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68cbce9abcff914079a7eeb3e7962ab64ab820db85e5207351f3512d521af2f0 +size 113246208 diff --git a/params_shard_141.bin b/params_shard_141.bin new file mode 100644 index 0000000000000000000000000000000000000000..886da387c09d350928a86b86bd4b71a4404c0028 --- /dev/null +++ b/params_shard_141.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b053d85d458e86aea3c3f3a95eb111687bc585e0b32566b416c91f40cd34cdb8 +size 25999360 diff --git a/params_shard_143.bin b/params_shard_143.bin new file mode 100644 index 0000000000000000000000000000000000000000..2c44d7fcadf41c1f6258c3eb5b2c57fd301026d0 --- /dev/null +++ b/params_shard_143.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:307f17c48fc8443653ea2630461cb364a8dfa031eae95cd32b9d4e3e60a9247a +size 56688640 diff --git a/params_shard_147.bin b/params_shard_147.bin new file mode 100644 index 0000000000000000000000000000000000000000..ec539a814f95b67ab4031d71d5d97f655410434b --- /dev/null +++ b/params_shard_147.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e443359691b2bed654b00827ad8c2ad5553464dea10696b5f0c926ba7c72d46 +size 28321792 diff --git a/params_shard_149.bin b/params_shard_149.bin new file mode 100644 index 0000000000000000000000000000000000000000..982e5a85550daa560ef7eba68e1da6f57a070e34 --- /dev/null +++ b/params_shard_149.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:071a7ab848926387bf78123ba2b382c648abd39ccd046b5958365f9578b4d294 +size 21266432 diff --git a/params_shard_155.bin b/params_shard_155.bin new file mode 100644 index 0000000000000000000000000000000000000000..ba518a9bbbb116543babf5800d84abde6f8588c3 --- /dev/null +++ b/params_shard_155.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba8a0a3f8079d27ba1911887540fa7f04eeadf6c29040498b9ee2620885a5ed9 +size 28321792 diff --git a/params_shard_158.bin b/params_shard_158.bin new file mode 100644 index 0000000000000000000000000000000000000000..82ff9a75d36f390c92fb0ff90ca4adf349c7bb14 --- /dev/null +++ b/params_shard_158.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0bff37d71233d09e2dec2d24324165b7b3818b9d9860d703016417db6979803 +size 56688640 diff --git a/params_shard_16.bin b/params_shard_16.bin new file mode 100644 index 0000000000000000000000000000000000000000..e30ce9dd9c7ca80883596d534dd74b3e88ae8f12 --- /dev/null +++ b/params_shard_16.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed28031ff619cc3fceb8ebc230b36ba5206077b63611a93bdb3d0d8f9befe172 +size 28321792 diff --git a/params_shard_165.bin b/params_shard_165.bin new file mode 100644 index 0000000000000000000000000000000000000000..f514f7433b2c47de69a1ec11e501857497f32b37 --- /dev/null +++ b/params_shard_165.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d4de65e5eda8229d4f1db2a3964546a7cd3db5f7aff8097e4294bad074fefed +size 21266432 diff --git a/params_shard_168.bin b/params_shard_168.bin new file mode 100644 index 0000000000000000000000000000000000000000..4b579d44c120670ded73fc2b697bd37f889ce48b --- /dev/null +++ b/params_shard_168.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c13138ed4347e651a5376d2fa42b1abe7df31c96d80445c56181ebc6e6bc9242 +size 113246208 diff --git a/params_shard_177.bin b/params_shard_177.bin new file mode 100644 index 0000000000000000000000000000000000000000..51a4281e3c479d79928a7b827a4edb167e7f3a28 --- /dev/null +++ b/params_shard_177.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c24a2339345bc4ae87dfaff244055d4f1152e81cae96a859b19db1b6ba42c03 +size 56688640 diff --git a/params_shard_179.bin b/params_shard_179.bin new file mode 100644 index 0000000000000000000000000000000000000000..19e830bee83f20e61514750c5056a3f6ee178cc5 --- /dev/null +++ b/params_shard_179.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cabcf55d70e08393f3c87a356263ca16ff09622f6b5a49f2cdc40f0a907f100e +size 113246208 diff --git a/params_shard_18.bin b/params_shard_18.bin new file mode 100644 index 0000000000000000000000000000000000000000..2d9189cf79673eb574991b3a4ecc863925e8c570 --- /dev/null +++ b/params_shard_18.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df6690269ed4b715d1db6428fc95b516c09a21e8b041420e0ce06e9f7907832c +size 21266432 diff --git a/params_shard_180.bin b/params_shard_180.bin new file mode 100644 index 0000000000000000000000000000000000000000..e8c826441ca39ca72bd899208a2d6b37da0846fa --- /dev/null +++ b/params_shard_180.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b14022ea491a1eaf888d9d8219d8a06837c6447ed19fd23a5d3df5c9db462dce +size 25999360 diff --git a/params_shard_182.bin b/params_shard_182.bin new file mode 100644 index 0000000000000000000000000000000000000000..ecb96a413eaf96f4e6214a46520ab881c9753b9f --- /dev/null +++ b/params_shard_182.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac94acbec540f5c61c70433acd05c56f2efb143e4a86723996867ecb8568663d +size 56688640 diff --git a/params_shard_19.bin b/params_shard_19.bin new file mode 100644 index 0000000000000000000000000000000000000000..327a617ae527d1566ca81a8dc0cca62303f1e672 --- /dev/null +++ b/params_shard_19.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:142107c860df2083a0e47b47093ea31921116e12a57f52bf166ab92c4328446b +size 56688640 diff --git a/params_shard_191.bin b/params_shard_191.bin new file mode 100644 index 0000000000000000000000000000000000000000..447391861e074a7a5e2df66fbc3c53212e0cf3cf --- /dev/null +++ b/params_shard_191.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3eb2b75e418c25f4c085d220978cea2b3b44a70a24e0aceecaf4794f5d0a708d +size 113246208 diff --git a/params_shard_193.bin b/params_shard_193.bin new file mode 100644 index 0000000000000000000000000000000000000000..e37f956408eece6b53f91b5248cfc0af6c164473 --- /dev/null +++ b/params_shard_193.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:966b63f954dc4ca31b038c94810e8051f65aae80b21ff6a1fb1036362f4f95c7 +size 56688640 diff --git a/params_shard_196.bin b/params_shard_196.bin new file mode 100644 index 0000000000000000000000000000000000000000..3ce53fa1be748b6adf0fc41845cb062d428cfe0d --- /dev/null +++ b/params_shard_196.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15757377bb91af4d1c267e0f258809a2d3b56ff421231fb3260f9fdcf5133998 +size 21266432 diff --git a/params_shard_197.bin b/params_shard_197.bin new file mode 100644 index 0000000000000000000000000000000000000000..6b7b85309a0863e97f9e1b4263ad7505e18898c3 --- /dev/null +++ b/params_shard_197.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a4da74f7195d9b82e56760d160be71bcf85fbd54f2b262136f8638b9916137d +size 56688640 diff --git a/params_shard_198.bin b/params_shard_198.bin new file mode 100644 index 0000000000000000000000000000000000000000..98ea82c185a56fba821ae15124fcf056e9921b24 --- /dev/null +++ b/params_shard_198.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69ba4516ed3deff4345993afa51eb87ab739221b929902a8c71a6366d3f9d331 +size 28321792 diff --git a/params_shard_202.bin b/params_shard_202.bin new file mode 100644 index 0000000000000000000000000000000000000000..0ca795cdd25ab9f6fd95c507db1e09712dbb82e4 --- /dev/null +++ b/params_shard_202.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfdc539c9a7cafc2e054ec197109cfb7eece75cdd73a3e44ddfb77b3ff30cf11 +size 28321792 diff --git a/params_shard_205.bin b/params_shard_205.bin new file mode 100644 index 0000000000000000000000000000000000000000..29340b24884a9929dac299de30a76031002f6797 --- /dev/null +++ b/params_shard_205.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8733160224ded0c71f5fc17538247216aae636d23187a4dffcb477a6253b9269 +size 56688640 diff --git a/params_shard_211.bin b/params_shard_211.bin new file mode 100644 index 0000000000000000000000000000000000000000..ccd306ba6faf526e6bc80cf114c81eed7b9b701c --- /dev/null +++ b/params_shard_211.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f6e2932c334104c22943b2adb4160b685a3cfe461ef1ce064e6b6c15fd21f54 +size 113246208 diff --git a/params_shard_212.bin b/params_shard_212.bin new file mode 100644 index 0000000000000000000000000000000000000000..f50f2d84a73a1b9db8e895637445cb450974d045 --- /dev/null +++ b/params_shard_212.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:024c91449d8af26d9e96961d7631ea022c8d26bd796093bcbfb559fe50746bda +size 21266432 diff --git a/params_shard_214.bin b/params_shard_214.bin new file mode 100644 index 0000000000000000000000000000000000000000..b5deb161ef82ae38287f96c0e9a701c84e530510 --- /dev/null +++ b/params_shard_214.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ea17b5472d4aebf0d50114318bf0de0942ce77918fa21b173724675bec3cd9e +size 28311552 diff --git a/params_shard_215.bin b/params_shard_215.bin new file mode 100644 index 0000000000000000000000000000000000000000..3336854bfee718df513a2e6e1a98f667cc9e9f30 --- /dev/null +++ b/params_shard_215.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a2754846e6c035fa108495cd8430fe78465fb451ee9c6d017117409b7bc07f4 +size 30685184 diff --git a/params_shard_217.bin b/params_shard_217.bin new file mode 100644 index 0000000000000000000000000000000000000000..f5b2551c2f739ff0fc910468b77eaf90f9639a06 --- /dev/null +++ b/params_shard_217.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb0e096958d1c388ce7a38162fddc113732e86748a6e311866f9089faa5144b1 +size 56688640 diff --git a/params_shard_223.bin b/params_shard_223.bin new file mode 100644 index 0000000000000000000000000000000000000000..bc8ce692452cd0c9d887958b16440d09a25c7e18 --- /dev/null +++ b/params_shard_223.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b55537feda6f68ba1f1cc451840b634beda45dfed53a1db85b8105c659f4e4a +size 33073152 diff --git a/params_shard_245.bin b/params_shard_245.bin new file mode 100644 index 0000000000000000000000000000000000000000..c0f20d0eb9bd207e487b014cfed7a2d5c0cfab79 --- /dev/null +++ b/params_shard_245.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b62a35ef200fd73593e243b25d52dfae5dd1d4ad8077355a45802bc19ca509eb +size 28321792 diff --git a/params_shard_248.bin b/params_shard_248.bin new file mode 100644 index 0000000000000000000000000000000000000000..339699f87702a06f5958f0c04755adeb9119ef94 --- /dev/null +++ b/params_shard_248.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b27f8a94a5e043933709ca64acb12938a4f6060b53e6cd697b8dd88867b8d91 +size 56688640 diff --git a/params_shard_25.bin b/params_shard_25.bin new file mode 100644 index 0000000000000000000000000000000000000000..2f3936b7ddee5522e7d8a64773098e2fb88bf963 --- /dev/null +++ b/params_shard_25.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e3358a00d20e74e1205b62363ffbae04cf76a650bc89f0b7d0ac502820184e0 +size 56688640 diff --git a/params_shard_253.bin b/params_shard_253.bin new file mode 100644 index 0000000000000000000000000000000000000000..292f0ff6fb98764155cc9ac2df9825033f32133d --- /dev/null +++ b/params_shard_253.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e566c319d725a5afd4d4ed21964a992a58fdc8ac335ff1eb219d3e5c642b261 +size 28311552 diff --git a/params_shard_26.bin b/params_shard_26.bin new file mode 100644 index 0000000000000000000000000000000000000000..a4ea056d6a0dc82910cfe813bc5f686d498550f6 --- /dev/null +++ b/params_shard_26.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2df0684d63b0d2de3c41a791bb104b69b5465222699869a862951c52bb9904dd +size 28332032 diff --git a/params_shard_27.bin b/params_shard_27.bin new file mode 100644 index 0000000000000000000000000000000000000000..b748e476db8be227bf2d079621aec400e39be13a --- /dev/null +++ b/params_shard_27.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1b9c71bc239ed55f3b0e9d16f37fdd7fd1b9c4184d495d52d874cc1813e973e +size 113246208 diff --git a/params_shard_29.bin b/params_shard_29.bin new file mode 100644 index 0000000000000000000000000000000000000000..eaddb948b6a156a90d16afa5325e863f0d60c89b --- /dev/null +++ b/params_shard_29.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:472742a63dd51916eff0c239a9444a6b2cd3df64787bcc4d2a2a90b761816f28 +size 56688640 diff --git a/params_shard_41.bin b/params_shard_41.bin new file mode 100644 index 0000000000000000000000000000000000000000..b2c961056d9bd804e9086ad017078c315b7acb43 --- /dev/null +++ b/params_shard_41.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30f003a5a9481a1826f0d3bada4ac72d0241ad28813173be23b629ec9234b27a +size 56688640 diff --git a/params_shard_46.bin b/params_shard_46.bin new file mode 100644 index 0000000000000000000000000000000000000000..9b42522b6358844224784425f058b87edc7ed320 --- /dev/null +++ b/params_shard_46.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73fe1fde32b5f4f1adfec84c9c8a198df5f792a868b3822a07163ea4ebaed487 +size 28321792 diff --git a/params_shard_49.bin b/params_shard_49.bin new file mode 100644 index 0000000000000000000000000000000000000000..3747b21781b5c105fe312363033933cb268645c7 --- /dev/null +++ b/params_shard_49.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a53311f11c164bf60497aaf735d8b18272b4cfbaceaf1cd649d1fb1487312001 +size 56688640 diff --git a/params_shard_51.bin b/params_shard_51.bin new file mode 100644 index 0000000000000000000000000000000000000000..665e89c3657628d33e78b112be06e6fbefae0def --- /dev/null +++ b/params_shard_51.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9db02899b2bbc3bc5e588d5dde22ec8caa1fcad8c5fd43f767fc87abf81ae1df +size 113246208 diff --git a/params_shard_59.bin b/params_shard_59.bin new file mode 100644 index 0000000000000000000000000000000000000000..3b25d472bb06106ded0729db576ccf2deb12987d --- /dev/null +++ b/params_shard_59.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9df989e74593257a92f1c7f5923fe4814baca71195f9f7e44069a4ff4d231d47 +size 30685184 diff --git a/params_shard_61.bin b/params_shard_61.bin new file mode 100644 index 0000000000000000000000000000000000000000..f3c581040aee21c0603a028742044c6b9be93951 --- /dev/null +++ b/params_shard_61.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a055f257effd0dc02ab009db355eff88785f09634bc9625d81555bdb475e5376 +size 113246208 diff --git a/params_shard_67.bin b/params_shard_67.bin new file mode 100644 index 0000000000000000000000000000000000000000..3684b4a166f50412b0a02f0dc6010089b8382f78 --- /dev/null +++ b/params_shard_67.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa96293a212c8405bde9e78f3c8bc5329249f440e06e006c9777b27d4f81589e +size 56688640 diff --git a/params_shard_7.bin b/params_shard_7.bin new file mode 100644 index 0000000000000000000000000000000000000000..5f7579cf313848bb3c03076d4b95a7bbbb2cd8c1 --- /dev/null +++ b/params_shard_7.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97964d65d32aa07d190aac4f7e1d2b30ed391ceff860a0caf451b063d3dae549 +size 30695424 diff --git a/params_shard_70.bin b/params_shard_70.bin new file mode 100644 index 0000000000000000000000000000000000000000..eb90cd1381004370f825e079334bda09fec465af --- /dev/null +++ b/params_shard_70.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c44a59cadadd3b258cbbcfed9603facf3c0bf4f6ff37939befd47704fe2deeff +size 21266432 diff --git a/params_shard_71.bin b/params_shard_71.bin new file mode 100644 index 0000000000000000000000000000000000000000..e7e0633db0e5de342e2bd403a8e58f36fa10997b --- /dev/null +++ b/params_shard_71.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c05cace3dd75a05ae55dc9c1c40da069201f48bf62c169e1d97d82f15494444 +size 56688640 diff --git a/params_shard_73.bin b/params_shard_73.bin new file mode 100644 index 0000000000000000000000000000000000000000..c38a507ec9bb4ededcd3a86010bf90d1811f5635 --- /dev/null +++ b/params_shard_73.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:276c63468413a0fcf10921312ad10a9c6b3366a60910ef82017f497e7da91f7a +size 113246208 diff --git a/params_shard_75.bin b/params_shard_75.bin new file mode 100644 index 0000000000000000000000000000000000000000..14c6deaae2d8291db5f7830e3f4cf39266f3a499 --- /dev/null +++ b/params_shard_75.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a95c2b1403e80607c6d58235f969ecba0f6a3a30aa306ed33f154627c815a094 +size 56688640 diff --git a/params_shard_78.bin b/params_shard_78.bin new file mode 100644 index 0000000000000000000000000000000000000000..9051863666c58877582fdc21f997192d33890e7b --- /dev/null +++ b/params_shard_78.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1a534bd67b2253defddf901b04be55b4908cb9b69295033ed4c1580359b0485 +size 21266432 diff --git a/params_shard_79.bin b/params_shard_79.bin new file mode 100644 index 0000000000000000000000000000000000000000..79a4c5d276a35323a4bf33d8d52dacff8fb46480 --- /dev/null +++ b/params_shard_79.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dee19593f467ec1421eb5f86aa2ffa3c94474b40b5e66c869578e334a74a8d8a +size 56688640 diff --git a/params_shard_84.bin b/params_shard_84.bin new file mode 100644 index 0000000000000000000000000000000000000000..98c9469260b3a0ad234fbeee0bf876803024b51a --- /dev/null +++ b/params_shard_84.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:04a54afde526522b88108c81e1016b1c9fd0736d55e044a4cf62f24e31c59a1b +size 28321792 diff --git a/params_shard_87.bin b/params_shard_87.bin new file mode 100644 index 0000000000000000000000000000000000000000..09ed405dfd594114f67edc440234f357377f7dd2 --- /dev/null +++ b/params_shard_87.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:757f5a9fc89ee8606dfe1ce68910e8e103d09e8f70c76b668e19d981079f7dfe +size 56688640 diff --git a/params_shard_90.bin b/params_shard_90.bin new file mode 100644 index 0000000000000000000000000000000000000000..ffcc47cc1c06f91600a1e982d1cfdc2e1ecf2a23 --- /dev/null +++ b/params_shard_90.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba252af5ad5ccff040f2deae49b188c2b55526995b0af1b7b3abaeaf3ccf61db +size 21266432 diff --git a/params_shard_93.bin b/params_shard_93.bin new file mode 100644 index 0000000000000000000000000000000000000000..15118317464f391ab8c2d16ee6ed73196c1b29fb --- /dev/null +++ b/params_shard_93.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b0a8e91cae46d54b9c2b1c91007bdce0d45d658c1adad06dfd3ffc65cfde579 +size 30685184 diff --git a/params_shard_96.bin b/params_shard_96.bin new file mode 100644 index 0000000000000000000000000000000000000000..6f2c6249bda54c529669e3cbe49c803e534bea27 --- /dev/null +++ b/params_shard_96.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3c3c55a5ea72cbe9ddafce54a3ee4a63cc3a8efff918f36d02b5adf3c59e5d4 +size 113246208