diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4175 @@ +{ + "metadata": { + "ParamSize": 305, + "ParamBytes": 3887439872.0, + "BitsPerParam": 4.5004158020177805 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 209715200, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 102400, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 209715200, + "byteOffset": 0 + } + ], + "md5sum": "4ab27a578d52f054b2e8b37650fdf951" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1fb89142af5f969230827ac4f9d74b70" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "30ec0faf19b6c25b14e8850bc52af747" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 29040640, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 102400, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 26214400, + "byteOffset": 0 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 26214400 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 26222592 + } + ], + "md5sum": "b694bd028d197c303feda9d5ab028462" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "1ebfa616cf1cfd37254d9e8e22ff3d11" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 31014912, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 5636096 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 5644288 + }, + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 5652480 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 28196864 + } + ], + "md5sum": "e5b9cdbcda60e82924469509a66f3204" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 30810112, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 5636096 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 5644288 + } + ], + "md5sum": "ef0b0d98afa546a369737c77d9b529f1" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dea9a6c0ec764fa59c2d7e0deebc2506" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "debdf4428bd68ac73ee6b7b15bf6886e" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "715c1c07e26d3265051c4b411739cd78" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "ee63f3b1dfbdaaaf511028894b620e9d" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3bb818ff51618b6bf78cef5ead34179b" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d7c576489e772e0f05d63bad9f96ecea" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "6d0cb8e07b6586cd16e30224f26ba303" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "86e987576a5eba111d46dc98d1a4cffd" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3ecf1c480aa05188bc0aa887f1431dd2" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "b266e08e2bd9567ca2c0c9a81eab8deb" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "e39591efda4704ca6ebf8b11ac057deb" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "95b5d1e3014543b6d400d1c294786cff" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "8592696f3e8d279a760d035c65a45780" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "267feee568aa926c9a19cebf0dc55616" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "a79e12e82c46a406c639b8669c80b90c" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "2965a5e3c5e620d19e186b94bbffc69d" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c74442e4d95504315c75081db8c9ffde" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "8ffed279d3f4ce0e907dce1687c6911f" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "a6e06fd5794b9c6bc5b22289610f679c" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "a4385a0fb712447176b18441d4792a72" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "487737d358af440dbe1e5563683dc02e" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 209715200, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 102400, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 209715200, + "byteOffset": 0 + } + ], + "md5sum": "7442b0f8c792f4a503e7426c2eefe00f" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 26214400, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 102400, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 26214400, + "byteOffset": 0 + } + ], + "md5sum": "b921813ad4cf34f77c3537d382aed707" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "960044f2518df8e9ca9a90d9d5a75fd0" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "273573a497c6703bcd96dfdb258c8fff" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "5d0c13946469ad89f1446ec912c9090c" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 32595968, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12591104 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12599296 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15417344 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21053440 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21061632 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24207360 + } + ], + "md5sum": "4c7e0ef03de99dd560e92d28c43e6b8f" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "1a5c37a4358206a2aaf0f57e17beb5d2" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "eff63bbed639ca436a010744b79316d3" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "bb58ff2cdfbea1fd9fedef12917231f1" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "8dd192b9fef29470b6ccedbb7e8927a0" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "fd937aaf1c410f9c908bcee12c217072" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "e33bb55002fe4331890935693510c44a" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "31f2d5ed1a5efc2050cce837ac1db2c6" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "33f7007b837f8be4ece794b636f4cfd2" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "8be34205c6c2fe89fb0f43c322f61a13" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "4c8cf8da245ee1083fa5d5936b122977" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1c741ce85e684052b80cf1aaa0b4a216" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "7884a145679f4159f4341e4cf8ad4f63" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "68ddab2ba5e638e8792e0bb38335d761" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "0c7dee8264b8ec68473b7541aeb8c7ec" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "ebfa2a39bf22857853df2e5165b67953" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d08f27985b097f2d7b62bf937c509dcb" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "c2a19b1f4aef4269053ab7cdeb1f64ca" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "1863faf2918fe49b1f35053d74bff400" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "55901832b4ecaa69a5900664d32534de" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d220a5c5cf8ac40d05f8066f5917a896" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "40f336e5f47d1d0996435fc7e0a953b3" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "08df47f65c23036dd4807286d85b28da" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "b04745e3461396b3c7be9c93c31c8a34" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "fd0fe24b14d80c5cc9e6f70a9df46d07" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "dcb23cd006a9b2c81dd029104e07a161" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "42e9adaed778411070fb7c0712bb25e0" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "1063e5102189df768500249bc2be8d9f" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "152498056aecd6c521cefa7b58c31eec" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "ed82185031c55d7c9d0f017cea162803" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "8466241a7d76a704b8fe70218011ca1d" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "2357a2bebccf37f0f2f89c2fdeef1837" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "72cd8a5a8df8d4999fad32e32921ea86" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "6c84c8268c72e0a683e98e5dc53bfc11" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "6df9dc899bd147bc7971c844072b46c0" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "dd726ac5f4dddf5b57907570d92fdbfd" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "267bfe7d16d1f002cda425aa5e188df7" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "931461a927476894744073e3d5a85ea4" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "1da941b414eea39bc12718b5cacc15fe" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "7601b8552d27ba890cc00db95f72e532" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "63d62493c08cdd6a71682323851180e8" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "42c718e6ee6b92d256a89f9f4b6c88dc" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "c5059dc9f3594aa423d2938b0e458f88" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "a96beb621128435c353130454eb8cbd6" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "ed434a456954dde9bdd6565a1e51c35c" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "95ac5972c399f43c43986912fc0b0c55" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "170027a321f9473ee9d9ecb6087334c5" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "65f70242f53394f944eaf8c4f09a9c92" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "fdab0d8fb60a4c4573d8552203644fb1" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "d59019e14c4e610fe50b6e19fcfd07aa" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 1048576 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 26214400 + } + ], + "md5sum": "caf447dccc6382281c1ce5bfc892fdbc" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "0814d71bfe279293a52b69076bf98bd4" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "c82ade38e6d558684fcedf998d41f4cc" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "bd312b26e441335593acf56aa7ae0083" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "02efd81ed4df52ea1c6b6da4bad4d73f" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "602a5f809089f7f5908a63247c4a2931" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "8f8337d6740fa20aae45a52b9e6cc910" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "d59ebfc3c3669d6d6c6d2400c4831ac8" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "97e1e4d8fb4af57df817fc4c88e46ca6" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "5a66a137486f43b1d02dfac84c61909b" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "ede86eef461fe236eecfb6b5c7e41e94" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "30b90dd26596f3dba4f326b54d7cffad" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "7e59f7e822433f17e863020de24cf858" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c29ce99bbbf076a68f1f1f8bc71f3fe9" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "7f0b9cece8b9a4173121a183284142c5" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "1b6f8ea48500cdb251f91a38c6ade5f6" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3e39ac77df99010cf712b4cce480a9d9" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "3bfcb8049c1643113ab0ac220641f5f1" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "93bd687deff69992924ee95d58d64b59" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3827b1a42e4775bfd20dfcac47f48e57" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c9ab122ebbfd820cef4fe016fab053b3" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "9fe2d0cc8c03ac9d44b562dc6cff8483" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "87d28fa973f0e991ed820405fb26c9f4" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "4067c6f923a527737d3e0444d4c6d0f4" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "efb0124c0bc49832a72acbd756715b69" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 21045248, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + } + ], + "md5sum": "0c5377bd78006cfb36ea00dc7f2bed36" + } + ] +} \ No newline at end of file