diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4303 @@ +{ + "metadata": { + "ParamSize": 325, + "ParamBytes": 3734421760.0, + "BitsPerParam": 4.125436579647682 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 65540096, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 32002, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 65540096, + "byteOffset": 0 + } + ], + "md5sum": "a2dcbae1b9da5dc0a6c3c0d92204e084" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "83d236bafe1f4fe2df57ee3e59210e8a" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 33267840, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 32002, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2048128, + "byteOffset": 0 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2048128 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 2056320 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 3891328 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 3899520 + }, + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 3907712 + } + ], + "md5sum": "48e3f5aaee401df406fe8b5d5cf02340" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d24919297c1578f08573f50253fe4fe9" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f5ec61db27a552416027d179fef9f6c3" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "4bd600247c94f2a07e2e95dc5ec41b41" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 27156480, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 0 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 917504 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2752512 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 2760704 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 15343616 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 15736832 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 24125440 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 24395776 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 25313280 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 27148288 + } + ], + "md5sum": "b1329eda509598a774fd4ffa8ccdd29e" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "ddaaba3dd2ed60ee094e7682c2b8f583" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "ff13def09c82fd9016e9ef77075c762f" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "9d3bfa844f96298ecc8e84df26c517c3" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9292f48c3348c599d53782c349b4e617" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "63df8f737221b1cc9173a7ac474cd49f" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "0f0197ac9549f84d969cbaec0f0660ca" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "cb4b28bff6be68ca6893e928f3071333" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "bc4f8aa5d6ee3fe2901b288b60601e1c" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "147d5b047be05dbecfe132adc7e64813" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a4ea4ca3db831ee6304e2c02a8002713" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "45cca5221fe0d57a749fb26efcf96b83" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "6ef55cad6cf4f72306a52c5de6ce1c77" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "8a2878cd12ab54d6231891c069dc35ba" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "27618303ad03a70296c46204f9725e31" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "ef620e9ab6162bcf91e0cfde0560bb2d" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "68de916d90e822962bb31d946137bff4" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "60db7ee525b9499b86ab26100f08fab5" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "83c282f64c73925dce544e4fd0c6852e" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7ada513ebbab639fe17a6c4840841871" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "8dd322f5237ccf283b0ca57384a38287" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "3129e08f6435a5ce8e4c9bf2e20dde0e" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "88fef360fb0e484a5a3a201d99f9f7fd" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "cb7c069a499e444a63bf609c7a188bcb" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "e48ae4d277f0b432863a476c60c9efc6" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 65540096, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 32002, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 65540096, + "byteOffset": 0 + } + ], + "md5sum": "41d72cfa6a998ac42c933bf43736317a" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9e64acbc2d61b8ed5a3dd3a5a9169f45" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "62434b3ea3974556e447ee65787d4470" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 26452096, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.31.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 32002, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2048128, + "byteOffset": 21635072 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 23683200 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 23691392 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 24608896 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 26443904 + } + ], + "md5sum": "a1b48063008d27e101709f5006b1128f" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "19695f3f8aaba5fcf125612eb63805c9" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "268c624d8ee69b80747bb33083cb30a4" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "1984db4e2bb4168ac7a65c480b274d61" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d04caed45004c643f8d1a6d5235caaea" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 23461888, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 21626880 + } + ], + "md5sum": "c5a5e65b57e2903078cea7639cad59f3" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "03fe4d45b5cf91b06c3c2263c21eacfd" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "b4a7538eda60756accaa7cf4618e8f05" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "d165ad44d05dc0dc7ed4ff241759c4bd" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5618861549f5fcd09ed1d990cf300315" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "11efa9e49218f042fde3fe697e67ea85" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "096f2675cb591877f05aa9e05df32ca5" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "52629650bed64ec10a5ed703e49f964f" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "61c1743e9c2de3cdfdaba1e18cf46271" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "69010f229afb21730902412d384ec180" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "e66d0309b513e97dea61c68324123cae" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "a79b714c7517f425bd3eb56ee19403d0" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "c76c812a70ede4a292bdf386a47ab204" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "21e167050c9e0138b0e842204baf5cda" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "672494dac3786bd614aec3c2a7c029b8" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "cd65106c1e03eea4d13ca24a4480913b" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c6911a644e228972caa090b068f12dcd" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "24f73714b9da83336bebcda7d55949c0" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "e0639ef629ac1c2cec30d2b2086a7b00" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "aebb52c9079bf033c9e16cadd3fa3190" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "895ed5fabf987aef512a17fe2b6b1272" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "763c8aaf3a8763b5b8a9c247ca6a372b" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9b87be10614eeb9adca298400f881050" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d7219b4a29ad692f86591f88e35ab950" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "e96ead9b66405440997a0519ff39d442" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b8836e4db86947f3b6550d53ead4bc9b" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "47cc59797fa11e7373f0f1db4a2ada52" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "1a9976f5767df8b96af3a01209358023" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 25329664, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 22552576 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 22560768 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 22568960 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 23486464 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 25321472 + } + ], + "md5sum": "27848f08616a3cd909ebe1ec992c9272" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "58c8374ace36df6fdb5f10cc6d582311" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "29fe6c53bff6d5705480af000df0ead1" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "b52dd4e2e63daaf261d59ba810e43b02" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "fab048c7c9b1083dfb7e59c5a3962cbc" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "c2e74e61c3ea2634422e62cee2590d6d" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "0b9976a40db6424b4f81e024320717c0" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "242a037008ab1204fc40d4f7952b0e5a" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "e6d0f7b44d4b7ef692c2d549b80fc491" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "8647388f778b2873ffa8b9d0b8c7c36e" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "12302bbfe427e1469df507bcdb4dc116" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "cbe307e63a18a96e30ab0f80dd39dd84" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "e6fe0da3100611b3970a86159812f829" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "2fdbfc1430bd6fa6ebeb0ffed8c73d32" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "1bb49941227586f7fad8c2aa8b1638bb" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "126fb265d8f6539d906c9259becd3448" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5353ed99e1cd6cfa5444fb9126f92bd3" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "0d199d1394283d1db29155ed95efc958" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "8878cac4d8bd34004a5592abcc04f4cd" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "54a45fcfc0bb6d50a2913e3938154140" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d76e70e1f0d151791d3316b21378c8f5" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "9243ab251bafbbc27b1b6d24ac04f502" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "dab33c7cca7f7f2ee991673a9b17ee74" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "f18a1327483a207cc5e9917d5e04ac63" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "a51aabb9867b11936905257a6b4a03a9" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7dd081c9448894e3e4a255a51fae7ec6" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "68686552ab9d8babad7b7cce542d654b" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "879a759c941e8acb778beb20fab0acd9" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5329091d33c187d85ad5df711d023c16" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21626880 + } + ], + "md5sum": "85f6a1e37b4a8a750a31b73216d1b965" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 21626880, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + } + ], + "md5sum": "cc10962f06a5ccd4789d86fd011ef70c" + } + ] +} \ No newline at end of file