diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..52373fe24473b1aa44333d318f578ae6bf04b49b 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/mlc-chat-config.json b/mlc-chat-config.json new file mode 100644 index 0000000000000000000000000000000000000000..69f967b19ecfbfbcacb3b8d5edfef0b2de4a4763 --- /dev/null +++ b/mlc-chat-config.json @@ -0,0 +1,80 @@ +{ + "version": "0.1.0", + "model_type": "qwen2", + "quantization": "q4f16_1", + "model_config": { + "hidden_act": "silu", + "hidden_size": 2048, + "intermediate_size": 11008, + "num_attention_heads": 16, + "num_hidden_layers": 36, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_theta": 1000000.0, + "vocab_size": 151936, + "tie_word_embeddings": true, + "context_window_size": 32768, + "prefill_chunk_size": 8192, + "tensor_parallel_shards": 1, + "head_dim": 128, + "dtype": "float32", + "max_batch_size": 128 + }, + "vocab_size": 151936, + "active_vocab_size": 151665, + "context_window_size": 32768, + "sliding_window_size": -1, + "prefill_chunk_size": 8192, + "attention_sink_size": -1, + "tensor_parallel_shards": 1, + "pipeline_parallel_stages": 1, + "temperature": 1.0, + "presence_penalty": 0.0, + "frequency_penalty": 0.0, + "repetition_penalty": 1.0, + "top_p": 1.0, + "tokenizer_files": [ + "tokenizer.json", + "tokenizer_config.json" + ], + "tokenizer_info": { + "token_postproc_method": "byte_level", + "prepend_space_in_encode": false, + "strip_space_in_decode": false + }, + "conv_template": { + "name": "qwen2", + "system_template": "<|im_start|>system\n{system_message}<|im_end|>\n", + "system_message": "You are a helpful assistant.", + "system_prefix_token_ids": null, + "add_role_after_system_message": true, + "roles": { + "user": "<|im_start|>user", + "assistant": "<|im_start|>assistant" + }, + "role_templates": { + "user": "{user_message}", + "assistant": "{assistant_message}", + "tool": "{tool_message}" + }, + "messages": [], + "seps": [ + "<|im_end|>\n" + ], + "role_content_sep": "\n", + "role_empty_sep": "\n", + "stop_str": [ + "<|endoftext|>", + "<|im_end|>" + ], + "stop_token_ids": [ + 151643, + 151645 + ], + "function_string": "", + "use_function_calling": false + }, + "pad_token_id": 0, + "bos_token_id": 151643, + "eos_token_id": 151645 +} \ No newline at end of file diff --git a/ndarray-cache.json b/ndarray-cache.json new file mode 100644 index 0000000000000000000000000000000000000000..7f90b3b15668f6438166f0f9396f3f42374ba638 --- /dev/null +++ b/ndarray-cache.json @@ -0,0 +1,4785 @@ +{ + "metadata": { + "ParamSize": 399, + "ParamBytes": 1736187904.0, + "BitsPerParam": 4.500900580433049 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 155582464, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 151936, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 155582464, + "byteOffset": 0 + } + ], + "md5sum": "a8f329f6ab0b7bc4d3b142d0f990cc24" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "11621cd4b1fbd1a6bf28fa42866098cb" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 32133120, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 151936, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 19447808, + "byteOffset": 0 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19447808 + }, + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 19451904 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 30724096 + } + ], + "md5sum": "0ed8548cb381dec97a27922669579e92" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "18a0910b4e572b42c772684643d8879a" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "f9a226978881ae7ba6458975dd83628e" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "b1a02fe0851610a0a181efcae641d1a0" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c6b9efc12099b53746e9a3fb22109c90" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "43f810349fc34e73d49773a43b2691be" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c1c0edd7bae6cc9801a19f813f0a875b" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "06f69167e277ec320c3cafef5d258a94" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6c326b56d9e0aa2c1f94569d9807e8da" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "eb1ee7869d5aa07d2ef427eec7cffadf" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "9d72412a3a00300ce4ac2c543fbe912f" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c2c410aae1e01fab011410aaaa5b58d2" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "1a2b98459c3177cd2ab18b5c0242e975" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "50fe624cc1f74eb46e2af5fb3e07c673" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dd0d4af95ba1a50bcbb25647bd4f0e1a" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "902f9bdef24591ae5daa11e674598470" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "a11e2c9b321ec0a43d1f931023611e23" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "465f73217a90b4e70de377d28356174f" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dc3dee7770afcc741de2e14a6485c176" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1e91ae347581b376d1b8fcc57efe907e" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "82b4c0c1091a6c10cfa47711e4f5427c" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d4c090020fb9b076d3b822d5272d7c95" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "835f2859668a9a0ed975bb39e374b862" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "744ddbc88bb00ee30cf632ced4127195" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "52a8cd27173dd6f7d92633ddce93ea65" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "15591d7aa011814c02652c1ffd978cd6" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "51bfa386319627c050432bee01093cff" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "e421ab69552076c3757c3c4f06b9e777" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1f3dc306f78d06ba50f70e0b4a7d2382" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "95296fcd0bd9d11efb19a2fa9dc615a8" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "4a64864b48b84a48480fc5470393781d" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6a6beb18e7d0d836c68a1bbd716472dc" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "656e009e42fa4f192040154e5bda2279" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d6346d860e76804316b94b1c4fab1e2c" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d2ed695a602c395fcaf935bcb446a893" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "b976732eea5181a579d58505aac4701e" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "b9639b7f27e9f5897613e8016da0d0f2" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.28.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.29.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "d8aa855ba014d321df51b8745c61c9b5" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "78dd38a3d37a287219193ba99154668e" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "32237f73544e0f5571e12acf488c32d8" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 33498112, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20816896 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32089088 + } + ], + "md5sum": "cacb240673dcc11e0201d1943cc4e881" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "f8a54c6535ad0394d0bfb6e151326edc" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 28956672, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2818048 + }, + { + "name": "model.layers.30.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2823168 + }, + { + "name": "model.layers.30.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5444608 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5772288 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7869440 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8131584 + }, + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8135680 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19407872 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20816896 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23634944 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23639040 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23644160 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26265600 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26593280 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28690432 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28952576 + } + ], + "md5sum": "60340ed231c87302578920db11ba987b" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "785188cbd5bb969efec41aba675e9d5a" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "9c3b0db39639f529038d532ca1316cb7" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "1189ba33f356805ca674cc4d66d7b252" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "7934666397ac8d94ffebadb5f7632627" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "13359b38b80084dfe3a94f165207d246" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "194e6dd743ad70d3a9a2bf1c256dc28f" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "13fe31e9bca0aa05a16d9c61f0dec445" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "00089bd24775ee8edb298fc07d60f729" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "ebda644edd7838884be52a2629153a42" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 28968960, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8139776 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8143872 + }, + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8147968 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19420160 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20829184 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23647232 + }, + { + "name": "model.layers.31.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23651328 + }, + { + "name": "model.layers.31.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23656448 + }, + { + "name": "model.layers.31.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26277888 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26605568 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28702720 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28964864 + } + ], + "md5sum": "b8fdc87fe3346ea1d173881d0028e1a8" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "8873af6074eb630a9911add44dd02674" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "61ad27902bd88b5f0eb9ace6de4167da" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.32.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.32.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.32.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.32.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.32.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.33.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.33.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "2c1689c1472c171165bdc91a80927c45" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "f00b8595d196deee8535dff7f54a3e50" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.33.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.33.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.33.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.34.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.34.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.34.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.34.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.34.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.34.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "411020c5e6a50ccc7dc960c5b102a4cf" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d21e8213c6018382d33455d1b57b8d7d" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 20820992, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.35.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.35.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.35.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.35.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.35.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.norm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + } + ], + "md5sum": "311b6fe905078f1b4aef390048a2c168" + } + ] +} \ No newline at end of file diff --git a/params_shard_0.bin b/params_shard_0.bin new file mode 100644 index 0000000000000000000000000000000000000000..81a780aa467acc298312ec80892e6f993af5779e --- /dev/null +++ b/params_shard_0.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f73b50f7b797c488151839a46bf1349cd106bf06d053887a504aae36de304824 +size 155582464 diff --git a/params_shard_1.bin b/params_shard_1.bin new file mode 100644 index 0000000000000000000000000000000000000000..4068f2f579d28aacb07d06ee1408f336a225ec7f --- /dev/null +++ b/params_shard_1.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4985e0963052c23e49c31aba75b8dc8d2160ecde3c930fb3535c96ece60117b5 +size 22544384 diff --git a/params_shard_10.bin b/params_shard_10.bin new file mode 100644 index 0000000000000000000000000000000000000000..b70807f36b7ef64bd6bdf599d52b55551a24a19d --- /dev/null +++ b/params_shard_10.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7caf37c01c950e0cc2b6cb4f7f67978fd46365a930948377f3e01d5ab21e242 +size 22544384 diff --git a/params_shard_11.bin b/params_shard_11.bin new file mode 100644 index 0000000000000000000000000000000000000000..6861fc7d1066ebb21574cb68bc1cbcc518990bbf --- /dev/null +++ b/params_shard_11.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2dd55ab91ca0e1490ff7823614553307087c3f63a1f80d9701a4ca74784782ad +size 22544384 diff --git a/params_shard_12.bin b/params_shard_12.bin new file mode 100644 index 0000000000000000000000000000000000000000..709dbda34ddd0db23218b30aa6e71c967a7cefdb --- /dev/null +++ b/params_shard_12.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:757d94bbcb97835e11a23f5007c02da087aa80accbce677fce604e08f4a5967f +size 33502208 diff --git a/params_shard_13.bin b/params_shard_13.bin new file mode 100644 index 0000000000000000000000000000000000000000..499a494d8aa9c31f992648992793b5d9a5b14fec --- /dev/null +++ b/params_shard_13.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee7f07caf45713601e79129f379145614607023197fdd1967faafdc49e516db7 +size 22544384 diff --git a/params_shard_14.bin b/params_shard_14.bin new file mode 100644 index 0000000000000000000000000000000000000000..6e8a3669832f74a58b77ca16cf7611212ce6b810 --- /dev/null +++ b/params_shard_14.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa65a0e680860163f507107f1ef96d5bebb3412288c5e71a42a9eb2bb58ea59e +size 28960768 diff --git a/params_shard_15.bin b/params_shard_15.bin new file mode 100644 index 0000000000000000000000000000000000000000..9105afa597d5e17fb045db71b16e56625295f141 --- /dev/null +++ b/params_shard_15.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42352377b0c86ac96b2a4884e62dd1d4bb0281b7182981f755ae0f772a3d6112 +size 22544384 diff --git a/params_shard_16.bin b/params_shard_16.bin new file mode 100644 index 0000000000000000000000000000000000000000..7e297c9b06f9d54e7a7295dd34e365076fc47df5 --- /dev/null +++ b/params_shard_16.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77accf41ddc3327e14bcbfc2c4a9f6878e00d77e287983e4ee7b24c24ba98e89 +size 22544384 diff --git a/params_shard_17.bin b/params_shard_17.bin new file mode 100644 index 0000000000000000000000000000000000000000..27fac20aceefe94d7802b91aab6a34655d2bce82 --- /dev/null +++ b/params_shard_17.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14c58147a1de0fef597ce5566e65f1c1f78a9392694641e9c05a67dfd666055a +size 33502208 diff --git a/params_shard_18.bin b/params_shard_18.bin new file mode 100644 index 0000000000000000000000000000000000000000..c399d525325a6657e30d1f7b261d8de28c00ba2e --- /dev/null +++ b/params_shard_18.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62557f44f1194b96a8ae71d1d66375b43ee86957e167c847f2b613ce809d6744 +size 22544384 diff --git a/params_shard_19.bin b/params_shard_19.bin new file mode 100644 index 0000000000000000000000000000000000000000..a26306ec0b57624e6658bc6f7aedf6d8d321c35b --- /dev/null +++ b/params_shard_19.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83f7452462c49db461f05e4f138134d97d6fd5eb5bb63846d3699c67a3160e3d +size 28960768 diff --git a/params_shard_2.bin b/params_shard_2.bin new file mode 100644 index 0000000000000000000000000000000000000000..038d6bbd3c794b09f6d9ae940552f62db57f8a0b --- /dev/null +++ b/params_shard_2.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0da35b8a9feaed603d22d8e4d8bdf12128f1eef22fb6da2aa8157416fb930398 +size 32133120 diff --git a/params_shard_20.bin b/params_shard_20.bin new file mode 100644 index 0000000000000000000000000000000000000000..f2ae0408e9456d97c55ccca5f125b1f603405a0b --- /dev/null +++ b/params_shard_20.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28719db046760004f9e03166ff78631bfa5d65f44fc4a14c215eb8193ac7db14 +size 22544384 diff --git a/params_shard_21.bin b/params_shard_21.bin new file mode 100644 index 0000000000000000000000000000000000000000..02d985758e9bbf9ba979c5cc55839051c5112ce9 --- /dev/null +++ b/params_shard_21.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9aba1c0cd9444ab87ae6df0edcfc7d0a85ef233582112cabf1f2a343f9c66c0 +size 22544384 diff --git a/params_shard_22.bin b/params_shard_22.bin new file mode 100644 index 0000000000000000000000000000000000000000..d9ba6380aebc38ae09cef952c365f4662236fc24 --- /dev/null +++ b/params_shard_22.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91a8c860589c03d4db101449a7f99aaddc54d30b72c5545fff217a97e2b7b715 +size 33502208 diff --git a/params_shard_23.bin b/params_shard_23.bin new file mode 100644 index 0000000000000000000000000000000000000000..e4a331c75a609dfe17b4bbf28ef44f5e3ebf00ac --- /dev/null +++ b/params_shard_23.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ed6992c62017c6c609d37b281be959ac8da50cefb9f63558494b786a62d6db9 +size 22544384 diff --git a/params_shard_24.bin b/params_shard_24.bin new file mode 100644 index 0000000000000000000000000000000000000000..d27ebc2d1c3abda0def568f2402ebec6c3000dae --- /dev/null +++ b/params_shard_24.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d1f94624e5d6566d0777fcbb1b621a413f3d9f6cd2055547e44e66b58c13e04 +size 28960768 diff --git a/params_shard_25.bin b/params_shard_25.bin new file mode 100644 index 0000000000000000000000000000000000000000..b40e7f2e0c8a1e38b46259351e4654d1566d5732 --- /dev/null +++ b/params_shard_25.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d013cfc6de42c7d11c07f05dee02cd2904505e0f25fecd391387ac44a576ac8 +size 22544384 diff --git a/params_shard_26.bin b/params_shard_26.bin new file mode 100644 index 0000000000000000000000000000000000000000..5d4fe85a7f5ec74e95fe1f4c6820dff1d430eafb --- /dev/null +++ b/params_shard_26.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:435d2984fb105b8eaecaca4649f0a21db579f6412d733dbcd8ec67e22db62211 +size 22544384 diff --git a/params_shard_27.bin b/params_shard_27.bin new file mode 100644 index 0000000000000000000000000000000000000000..071e7414527bd87297cf61fa60714bbcc9cbe3ac --- /dev/null +++ b/params_shard_27.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c09589db46dabbc0f4048766a4bcd897926495b277eb37a5774964b0e73eb778 +size 33502208 diff --git a/params_shard_28.bin b/params_shard_28.bin new file mode 100644 index 0000000000000000000000000000000000000000..5390664ae15d4e85626ae2b8c1e7dd3571daf932 --- /dev/null +++ b/params_shard_28.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2ff89a0a1787af0b902c7b466067a82a6e0156e4c070b114062802fff998d83 +size 22544384 diff --git a/params_shard_29.bin b/params_shard_29.bin new file mode 100644 index 0000000000000000000000000000000000000000..d97eb2af702818f339ce16c98d0ce7c9f3602e98 --- /dev/null +++ b/params_shard_29.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59cabfd5ce5fb44a3087d37ba9092fe6f6f0cc31f1b8ad8dd1b210723074d116 +size 28960768 diff --git a/params_shard_3.bin b/params_shard_3.bin new file mode 100644 index 0000000000000000000000000000000000000000..d941c9e4a8bf72522e4e78840ca791d51848e9d4 --- /dev/null +++ b/params_shard_3.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:278d1c5bf424be2c3776b56adac2284824701c61daa40d5f5790ffafdac59d92 +size 22544384 diff --git a/params_shard_30.bin b/params_shard_30.bin new file mode 100644 index 0000000000000000000000000000000000000000..d9bb35066f0a1a0ecb4ef4a09851ffb7942ff6e7 --- /dev/null +++ b/params_shard_30.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:244ba8abe59a7bebb256b4683bcce9ef8760d84847fa0151237b6ab0d3dcd8b9 +size 22544384 diff --git a/params_shard_31.bin b/params_shard_31.bin new file mode 100644 index 0000000000000000000000000000000000000000..399285b10b9a45e8a5eda4c33fb9bcaafb245321 --- /dev/null +++ b/params_shard_31.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cdc0c03e627618f861903559a6ed1d0e5d1c0b41618a52a3d7cebe2399bb261 +size 22544384 diff --git a/params_shard_32.bin b/params_shard_32.bin new file mode 100644 index 0000000000000000000000000000000000000000..da42e8f0d529e9bb617f199404d8af28e0ef8ed5 --- /dev/null +++ b/params_shard_32.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b86d7738a6d53f4b09008dbbfa85e64fd0a5e7c32d6313dbe6cb0ea9d8911398 +size 33502208 diff --git a/params_shard_33.bin b/params_shard_33.bin new file mode 100644 index 0000000000000000000000000000000000000000..25ce375c2fa65b7bf1994f455aa8527421658838 --- /dev/null +++ b/params_shard_33.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ba11e8dfc84b98e11ad339d8480d4b00453d2b28213162bf3a321384b425695 +size 22544384 diff --git a/params_shard_34.bin b/params_shard_34.bin new file mode 100644 index 0000000000000000000000000000000000000000..b865a72d66d27a9e99f2933d73b52106133c2df9 --- /dev/null +++ b/params_shard_34.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d04b1e30cbf7b99c74c63536db488d717704734819342a2b3d8094e1a6ac5dd6 +size 28960768 diff --git a/params_shard_35.bin b/params_shard_35.bin new file mode 100644 index 0000000000000000000000000000000000000000..0d6f08e9edb3164a70b1caa2cdcd89752ee9ce2d --- /dev/null +++ b/params_shard_35.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9154ff682fd312d36b507e432a4bd974b475d1dc765efae7e01146abdf41e06 +size 22544384 diff --git a/params_shard_36.bin b/params_shard_36.bin new file mode 100644 index 0000000000000000000000000000000000000000..2ab147215c55a4178be2af0cb2d9236018b5aa16 --- /dev/null +++ b/params_shard_36.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85cb2c3a78cb35fa45db615ff509131cba554921b2af9f5cdb2be805db0e4a32 +size 22544384 diff --git a/params_shard_37.bin b/params_shard_37.bin new file mode 100644 index 0000000000000000000000000000000000000000..5e498f3a6dc48bc2485095074226e80a3cbfda2c --- /dev/null +++ b/params_shard_37.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:780da39f0f60a39cf0b9aee6ab3f3e27956773dd89ae33d52d8975d92d656eb7 +size 33502208 diff --git a/params_shard_38.bin b/params_shard_38.bin new file mode 100644 index 0000000000000000000000000000000000000000..b2f66c53fd710796629d88fba7d2ab6354600099 --- /dev/null +++ b/params_shard_38.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76aa0b5b8d04c56be54dbcc3525ceb1edcbca65a66a6f2ddae8c37ac256b1fcd +size 22544384 diff --git a/params_shard_39.bin b/params_shard_39.bin new file mode 100644 index 0000000000000000000000000000000000000000..1ccf0f90592f15d66501353ec8b5e86d06db626a --- /dev/null +++ b/params_shard_39.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a292210c0277f763d222848c7d5a199d87912c95e2548e9c74b43c8f966d01e +size 28960768 diff --git a/params_shard_4.bin b/params_shard_4.bin new file mode 100644 index 0000000000000000000000000000000000000000..2f824bdff7e0c13202f37ccc0efea7ff5cb81728 --- /dev/null +++ b/params_shard_4.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a580f48c47a61f2ce4000ce0e0feedc311705773528c28f2ccd6467c446bdf57 +size 28960768 diff --git a/params_shard_40.bin b/params_shard_40.bin new file mode 100644 index 0000000000000000000000000000000000000000..044232f13e74f614b8ba55089d697448e077dd83 --- /dev/null +++ b/params_shard_40.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f6acd3c85b6e44303ef37fcb21b413d15229e5bf69fb6301b292b84bdb3ac6b +size 22544384 diff --git a/params_shard_41.bin b/params_shard_41.bin new file mode 100644 index 0000000000000000000000000000000000000000..4864e31935d03e403335e80147aeb8d3c9ef8252 --- /dev/null +++ b/params_shard_41.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:010afefd6d9db790fe19127ce5ba5fa346107ed5efdc39cdb17fee820e3c6feb +size 22544384 diff --git a/params_shard_42.bin b/params_shard_42.bin new file mode 100644 index 0000000000000000000000000000000000000000..2e7e4d25acebe1c8f0180970367fe5a0282b7965 --- /dev/null +++ b/params_shard_42.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad3fa5c0752f4d6cd9b63620627beb798d76b77f83fdfe6058931e7a458e710a +size 33498112 diff --git a/params_shard_43.bin b/params_shard_43.bin new file mode 100644 index 0000000000000000000000000000000000000000..9cf43f2e74517915a509ebc6d6c03b248365e592 --- /dev/null +++ b/params_shard_43.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8939a6b4428838670dcf488a504d20a3254c8b391a1c33cf5af9f59ea3504542 +size 22544384 diff --git a/params_shard_44.bin b/params_shard_44.bin new file mode 100644 index 0000000000000000000000000000000000000000..4a1c4d2099e8f421db4ca6c884a2d5fa955801f1 --- /dev/null +++ b/params_shard_44.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d27addc5170b7fd88c5ebc48014ed3e38f1b3d9d9482ba56a583e271d90f19d +size 28956672 diff --git a/params_shard_45.bin b/params_shard_45.bin new file mode 100644 index 0000000000000000000000000000000000000000..81b1785946711d93c4261c2813cad88217d2adc1 --- /dev/null +++ b/params_shard_45.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc8f8e6fc0ee910fce5c0f969a164a8841789410d6313cb3386c2d8cc7403424 +size 22544384 diff --git a/params_shard_46.bin b/params_shard_46.bin new file mode 100644 index 0000000000000000000000000000000000000000..6d812075f1b6dd51debd9f55da4e631bdd6030a2 --- /dev/null +++ b/params_shard_46.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf2e7283be704ac9a725e01afa2fc74ddf685d58d4caaa68f0e308e677295266 +size 22544384 diff --git a/params_shard_47.bin b/params_shard_47.bin new file mode 100644 index 0000000000000000000000000000000000000000..e5633b2c0f477754b1275bada19dbfea8439a375 --- /dev/null +++ b/params_shard_47.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eba2fe9fec5aa457cb563374b9d380df2b93034ce26bf5cfb79acf136493f060 +size 33502208 diff --git a/params_shard_48.bin b/params_shard_48.bin new file mode 100644 index 0000000000000000000000000000000000000000..204a22b7b70742eee24092b13a98a134b577e683 --- /dev/null +++ b/params_shard_48.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61c10a2a1d77ed36cfb891fe130b6fbc524751ad80803e4806293e67c9ba764f +size 22544384 diff --git a/params_shard_49.bin b/params_shard_49.bin new file mode 100644 index 0000000000000000000000000000000000000000..e29f023e6b6ce05f03d00c3650223a3c850d04e9 --- /dev/null +++ b/params_shard_49.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50c8f575b3f7d0e819c64a094384fc7ad825cbce83c986e9218fe774a43e371e +size 28960768 diff --git a/params_shard_5.bin b/params_shard_5.bin new file mode 100644 index 0000000000000000000000000000000000000000..95b96f9c4bab225cb97b3eab95b6e27a80c5163c --- /dev/null +++ b/params_shard_5.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9c7654484d9cdc6b17002b41fb42e2a851800d0a8e588337fb9d2b6f29dba74 +size 22544384 diff --git a/params_shard_50.bin b/params_shard_50.bin new file mode 100644 index 0000000000000000000000000000000000000000..c68d0131ecab08bcfc272971ba7d2eede6b967f5 --- /dev/null +++ b/params_shard_50.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79933840f2f8521fae7db43463de7398e5bd121db4ed886e0be6d77f36196dcd +size 22544384 diff --git a/params_shard_51.bin b/params_shard_51.bin new file mode 100644 index 0000000000000000000000000000000000000000..2272d4bf37f592363a8dbbd893cf1f4e0763ea6b --- /dev/null +++ b/params_shard_51.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64c659740ffb8b98c28d7d22711a79bd386e0cabaf082a4b62449271e29ecbd7 +size 22544384 diff --git a/params_shard_52.bin b/params_shard_52.bin new file mode 100644 index 0000000000000000000000000000000000000000..893de1987fa783584c751b95d6c30cb49e31e3d0 --- /dev/null +++ b/params_shard_52.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a1ec13da1e4dc2787436be8c3983c1f37e3d48c496c3d4a02a6c87ecb374d1b +size 33502208 diff --git a/params_shard_53.bin b/params_shard_53.bin new file mode 100644 index 0000000000000000000000000000000000000000..95d28070351cd07456b0245c4d77cbf6adc66d99 --- /dev/null +++ b/params_shard_53.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54c9d691e76a5c0254fbeeb839fc6bc4d597e6fd2941f2d72a1e3415e553b839 +size 22544384 diff --git a/params_shard_54.bin b/params_shard_54.bin new file mode 100644 index 0000000000000000000000000000000000000000..27d78b44bd2207f8355769d53732ed8a6b860992 --- /dev/null +++ b/params_shard_54.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45d5bc5d27da43e0362fa6e7d9dba8d56d9152fa4d57c5e6586e6faef0283d7e +size 28968960 diff --git a/params_shard_55.bin b/params_shard_55.bin new file mode 100644 index 0000000000000000000000000000000000000000..1f650384df5199a117284b42497f97b87cb822e5 --- /dev/null +++ b/params_shard_55.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79e89da2bb01bc347201daf5793be76e4f87d594d765e267fdf9fa80b3c67416 +size 22544384 diff --git a/params_shard_56.bin b/params_shard_56.bin new file mode 100644 index 0000000000000000000000000000000000000000..a95d814bc289b0d5e8430773838d6d5103e7ec8d --- /dev/null +++ b/params_shard_56.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9fe7d836a8e94cbd344dcd94521483a6a0a49dcd63e5fb388903167074054fcb +size 22544384 diff --git a/params_shard_57.bin b/params_shard_57.bin new file mode 100644 index 0000000000000000000000000000000000000000..d6d9e6c9ff60a468e85c741017cc5bb941254bd2 --- /dev/null +++ b/params_shard_57.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0600502923fe549f4dbd2701a47cfb82aad62d87aa0b6f36307ce243949f3329 +size 33502208 diff --git a/params_shard_58.bin b/params_shard_58.bin new file mode 100644 index 0000000000000000000000000000000000000000..648d52f002c13fc70524825fed88fbc486bb3136 --- /dev/null +++ b/params_shard_58.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4705838b66e0edc3ca40c110cf3d1776fd213dc797395ccdbc5da05ffdbcba76 +size 22544384 diff --git a/params_shard_59.bin b/params_shard_59.bin new file mode 100644 index 0000000000000000000000000000000000000000..e64ac46579d7c1c8c03d8684782e354640594dc1 --- /dev/null +++ b/params_shard_59.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a91b58d359f96d3e4780631c956f83af24c3d3d139de83257ac31e3be7c167b +size 28960768 diff --git a/params_shard_6.bin b/params_shard_6.bin new file mode 100644 index 0000000000000000000000000000000000000000..12c5653296a28cf42f7e9238212321cacd8b785b --- /dev/null +++ b/params_shard_6.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d01a78f2ce778eb13f275584aa8e09112f16cc147e1b874239cd61cbf36f748 +size 22544384 diff --git a/params_shard_60.bin b/params_shard_60.bin new file mode 100644 index 0000000000000000000000000000000000000000..6b56e1edeb2d857238140376f04c2b14eec1fbd3 --- /dev/null +++ b/params_shard_60.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:36e9870ffdda2bef9463b4e8a3e196832869337c9f252ee2cd48fdda86c6f688 +size 22544384 diff --git a/params_shard_61.bin b/params_shard_61.bin new file mode 100644 index 0000000000000000000000000000000000000000..45451d18c32f997cf9d6e90561de8a62a0347803 --- /dev/null +++ b/params_shard_61.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d8583b4873fc22c1198578ce5de812dd9037efd0c2f09e8ce7bcca8b2a094ee8 +size 20820992 diff --git a/params_shard_7.bin b/params_shard_7.bin new file mode 100644 index 0000000000000000000000000000000000000000..f3a8b8719b6a4139aa096a9435ed71ae7f4b2756 --- /dev/null +++ b/params_shard_7.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a68a4732e5414d8f7003731241ff837deb83481ad19895e2bcb93ff66981932e +size 33502208 diff --git a/params_shard_8.bin b/params_shard_8.bin new file mode 100644 index 0000000000000000000000000000000000000000..0407710cfd31ab94f5875089d71d1baf44911808 --- /dev/null +++ b/params_shard_8.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:731cdf991d807ec7576d5b3cdf834009ce9dd0aa841dd4d9a867356132b36693 +size 22544384 diff --git a/params_shard_9.bin b/params_shard_9.bin new file mode 100644 index 0000000000000000000000000000000000000000..3a0de5ba3b10cad4afc68372d4d14ca70e393860 --- /dev/null +++ b/params_shard_9.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:871045438ae6c617e0fd437c4d4d4bbda8544d624481b4aac897008f701e3731 +size 28960768 diff --git a/tensor-cache.json b/tensor-cache.json new file mode 100644 index 0000000000000000000000000000000000000000..7f90b3b15668f6438166f0f9396f3f42374ba638 --- /dev/null +++ b/tensor-cache.json @@ -0,0 +1,4785 @@ +{ + "metadata": { + "ParamSize": 399, + "ParamBytes": 1736187904.0, + "BitsPerParam": 4.500900580433049 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 155582464, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 151936, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 155582464, + "byteOffset": 0 + } + ], + "md5sum": "a8f329f6ab0b7bc4d3b142d0f990cc24" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "11621cd4b1fbd1a6bf28fa42866098cb" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 32133120, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 151936, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 19447808, + "byteOffset": 0 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19447808 + }, + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 19451904 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 30724096 + } + ], + "md5sum": "0ed8548cb381dec97a27922669579e92" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "18a0910b4e572b42c772684643d8879a" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "f9a226978881ae7ba6458975dd83628e" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "b1a02fe0851610a0a181efcae641d1a0" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c6b9efc12099b53746e9a3fb22109c90" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "43f810349fc34e73d49773a43b2691be" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c1c0edd7bae6cc9801a19f813f0a875b" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "06f69167e277ec320c3cafef5d258a94" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6c326b56d9e0aa2c1f94569d9807e8da" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "eb1ee7869d5aa07d2ef427eec7cffadf" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "9d72412a3a00300ce4ac2c543fbe912f" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "c2c410aae1e01fab011410aaaa5b58d2" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "1a2b98459c3177cd2ab18b5c0242e975" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "50fe624cc1f74eb46e2af5fb3e07c673" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dd0d4af95ba1a50bcbb25647bd4f0e1a" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "902f9bdef24591ae5daa11e674598470" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "a11e2c9b321ec0a43d1f931023611e23" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "465f73217a90b4e70de377d28356174f" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dc3dee7770afcc741de2e14a6485c176" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1e91ae347581b376d1b8fcc57efe907e" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "82b4c0c1091a6c10cfa47711e4f5427c" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d4c090020fb9b076d3b822d5272d7c95" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "835f2859668a9a0ed975bb39e374b862" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "744ddbc88bb00ee30cf632ced4127195" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "52a8cd27173dd6f7d92633ddce93ea65" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "15591d7aa011814c02652c1ffd978cd6" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "51bfa386319627c050432bee01093cff" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "e421ab69552076c3757c3c4f06b9e777" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1f3dc306f78d06ba50f70e0b4a7d2382" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "95296fcd0bd9d11efb19a2fa9dc615a8" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "4a64864b48b84a48480fc5470393781d" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6a6beb18e7d0d836c68a1bbd716472dc" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "656e009e42fa4f192040154e5bda2279" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d6346d860e76804316b94b1c4fab1e2c" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d2ed695a602c395fcaf935bcb446a893" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "b976732eea5181a579d58505aac4701e" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "b9639b7f27e9f5897613e8016da0d0f2" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.28.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.28.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.29.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.29.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "d8aa855ba014d321df51b8745c61c9b5" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "78dd38a3d37a287219193ba99154668e" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "32237f73544e0f5571e12acf488c32d8" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 33498112, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20816896 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32089088 + } + ], + "md5sum": "cacb240673dcc11e0201d1943cc4e881" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "f8a54c6535ad0394d0bfb6e151326edc" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 28956672, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2818048 + }, + { + "name": "model.layers.30.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2823168 + }, + { + "name": "model.layers.30.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5444608 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5772288 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7869440 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8131584 + }, + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8135680 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19407872 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20816896 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23634944 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23639040 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23644160 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26265600 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26593280 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28690432 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28952576 + } + ], + "md5sum": "60340ed231c87302578920db11ba987b" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "785188cbd5bb969efec41aba675e9d5a" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "9c3b0db39639f529038d532ca1316cb7" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "1189ba33f356805ca674cc4d66d7b252" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "7934666397ac8d94ffebadb5f7632627" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "13359b38b80084dfe3a94f165207d246" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "194e6dd743ad70d3a9a2bf1c256dc28f" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "13fe31e9bca0aa05a16d9c61f0dec445" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "00089bd24775ee8edb298fc07d60f729" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "ebda644edd7838884be52a2629153a42" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 28968960, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8139776 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8143872 + }, + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8147968 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19420160 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20829184 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23647232 + }, + { + "name": "model.layers.31.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23651328 + }, + { + "name": "model.layers.31.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23656448 + }, + { + "name": "model.layers.31.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26277888 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26605568 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28702720 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28964864 + } + ], + "md5sum": "b8fdc87fe3346ea1d173881d0028e1a8" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "8873af6074eb630a9911add44dd02674" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "61ad27902bd88b5f0eb9ace6de4167da" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 33502208, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.32.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.32.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.32.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.32.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.32.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + }, + { + "name": "model.layers.33.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 20820992 + }, + { + "name": "model.layers.33.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 32093184 + } + ], + "md5sum": "2c1689c1472c171165bdc91a80927c45" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "f00b8595d196deee8535dff7f54a3e50" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 28960768, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.33.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 2822144 + }, + { + "name": "model.layers.33.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 2827264 + }, + { + "name": "model.layers.33.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 5448704 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 5776384 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 7873536 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 8135680 + }, + { + "name": "model.layers.34.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 8139776 + }, + { + "name": "model.layers.34.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 19411968 + }, + { + "name": "model.layers.34.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 20820992 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23639040 + }, + { + "name": "model.layers.34.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 23643136 + }, + { + "name": "model.layers.34.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 23648256 + }, + { + "name": "model.layers.34.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 26269696 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26597376 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28694528 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28956672 + } + ], + "md5sum": "411020c5e6a50ccc7dc960c5b102a4cf" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "d21e8213c6018382d33455d1b57b8d7d" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 20820992, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_weight", + "shape": [ + 2048, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.35.mlp.down_proj.q_scale", + "shape": [ + 2048, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.35.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12681216 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 15499264 + }, + { + "name": "model.layers.35.self_attn.c_attn.bias", + "shape": [ + 2560 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5120, + "byteOffset": 15503360 + }, + { + "name": "model.layers.35.self_attn.c_attn.q_weight", + "shape": [ + 2560, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2621440, + "byteOffset": 15508480 + }, + { + "name": "model.layers.35.self_attn.c_attn.q_scale", + "shape": [ + 2560, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 327680, + "byteOffset": 18129920 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 18457600 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 20554752 + }, + { + "name": "model.norm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 20816896 + } + ], + "md5sum": "311b6fe905078f1b4aef390048a2c168" + } + ] +} \ No newline at end of file diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..34510ff0037cd50428af467a17ead5a96140a32c --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0cf1c38eafcebe924b401509a844cd24f2942efe --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,31 @@ +{ + "add_prefix_space": false, + "backend": "tokenizers", + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "is_local": true, + "local_files_only": false, + "model_max_length": 131072, + "pad_token": "<|endoftext|>", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "tool_parser_type": "json_tools", + "unk_token": null +}