diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,3137 @@ +{ + "metadata": { + "ParamSize": 195, + "ParamBytes": 14496047104.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 268435456, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 32768, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 268435456, + "byteOffset": 0 + } + ], + "md5sum": "880a529a6b63f015a04c866168bb9dab" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a24dc3c4d74bf48371bd7259267b7b23" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "217a194ea7567f2c56cb076938587a49" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e653ffdde8c32e1dcc38fc42d720a0ad" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c6c807553b470c2bbf80110395c1e000" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "74e4148d92e83ad5884bda3445861bf0" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "eaf3bbc8594469ff8753086cc029714f" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "8d0812804c1047c06361b904a1ab66e0" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "01b0c64a32bfde188e1f47865f845e57" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "d98a07fd0cb7bdaf1f1f027e0a77d0f6" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "fd3a6e3e413b9419ed00265ec8d3f9f4" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "375bd8e6f24a3e5462e88ccee21e4904" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "012ba97c851affb6421ad90d20ef5b98" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "2a599bd94125c58dc1469b7336592b7e" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "1c5fd051a9a88ddd6e5d695fb98bc21d" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e61aab8de931adea0e5491b137552fe1" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "f5e25d782f13f4878989c7da13ed9143" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "a4e4f52c88bfef6d577e3ca62e95c5b0" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "15b32ca2867a1aedc2f2a2cb6921b4a5" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9edf4eac66a01940e8488919d2888cd1" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "67c09e99cd3337ca04fac521c466025a" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "e9cf34ee7b5a92608fe84e76f5833aac" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c870dcdf77e3b677a3aba72b545d6e5a" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6d2c57f28bd66b05383bc3b47124e721" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6eafc10f5cec12b34434efb20d64b21b" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "7a49ef2f5792c40e3a2b976f21d86fad" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "9dc96ce63f7cbd32b82d33337123c98c" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "dd600eca85eed96368f2af23d10c2fd4" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c57fe2167e2d5fc6bc2b18a120c41f8d" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "216f1c723ba1abbd75e4399e9c627aa6" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "afde7ef939bcdee59f75a4f93b5bc129" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e51a4bccfc431a067daa2e4d7ec23a4c" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c2d5a89e79e964a7b7eb453fb37cb70a" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "86c3825a14bcf55d84af90e70f8e8844" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.30.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "63f1c22291e14b34395de3440a30fd9e" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "1cdf0e9bbe3c7e26fca500f394966554" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "260a0727a882b0133adbb1698c4e30fb" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "59c1e515a0151e62f21584341493765c" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "92858414a4480a593f6efb8afd2a3bce" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 268435456, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 32768, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 268435456, + "byteOffset": 0 + } + ], + "md5sum": "261978090a9465a0d74a7e0c9ac6e207" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6b621183f6f243549c8317671d9c9806" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6c47542e54001ed7cccad2c5d16c6cea" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "73afb8d0033c3ba39344b44247b06c0a" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d35a66b324b81bb48f491dfae5b282da" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6a323e19031076f694afc9f9819469be" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "4b09b3618f7e9f7b93ba719652204248" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "4a69742776561d500afaa85a90120894" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "73d6ca6a8f09f034464c7ed99b422115" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d2ba22038c6ea96fcf7588301fe2be1c" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "b545dbaedc844ab50d6ab331c1050815" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "8bc831e41066eda82789b47f4fcfecf7" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f88427dfc9f2b169ae7d85382922d09f" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "9e3e3343fe456de169c600502ebd81f0" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "84903b9469efc0af1c9bf15343fff82a" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "303366fe88f0c01d04dd39eb984c0a2b" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4cef6176d6911141d2d678370a439b69" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0437ae0dab251d92f9ef49ba1b190544" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "c0ebe1d94b4ec69510387bc96a503591" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "9096563b59ae4254d8444a385925f9a4" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "3e66b70f80e24f4eed25133d5d7bdde5" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "57924642903936535897b40c04871ee1" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "92906a3c1dd1be029acdcf13aa232955" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "4086c15f26e41e593f88bd6b9863216a" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ca2fc27bdf058f579599f511443b5890" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "7790a586e054e192775ade5dc4c3460a" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "a3bd0ae6a0b5923635d103314bed275d" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "b1f30c1754b202f9ee96f209a95a3661" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9c1c66d0c1070086fa1b39d2af22e8de" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "45bcedff4bcd4c16496cde60870a70c3" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "a0bbb1da6306640611b148e63f3def38" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "3c1e1768468ba22e692f9be9f5701f4d" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a32b0db6ff39b1efd7787b7e594dad12" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0e5b37d59ab72adb77f0e326985fd432" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "f3540a9dba8992cfb2fd41593d88b620" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d38269dbbd5493742f1ab287c2879bad" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4448cb575c0bce9727db2e2cd55905a9" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "305531d6991544c0201bd81cbf074b9e" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "462ac5603a8d36402e475ffea8d76d47" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "7b318d0cb41526db1067454ef5b9129b" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9f5e3c2db1a5fe19687b5881dd2fdc3e" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "126fa67040c21c06ebe19bd4eac48b1b" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "ac919fc939bb3cd29a362357e44d1aa4" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "96ed562be37e21d1288d56bae0062d18" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "8c5d5939b91207cd4ff8e962f07c4687" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "c3949ae168a8209dd17ec479d3b34f01" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "8efb29ded9acfb7a0e81f2ad725850aa" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "24212a94d9458bd0e57f1b77471bc056" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "f13b4b1fa87e78850f4bbdf28f4c6816" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4edaf4595ab78ec9709e7deef2b40214" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "fed440c819b1795853aff0ea8035b2be" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "5e5e23ac9a6be1f9f7db9fb71b7520a2" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d6db3bb1adfda14df7d0a716a7a67763" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "1149fe189abaf027afda27aa0343167e" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "3e1a69e4213c7753fa6a62ac6779ef5f" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "f55ab062a9554989ad43fc68c58025bc" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "5dc8e30847c897731a3b2ac7ad76babc" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "147194abc8a528ec1da31c5de8dea841" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "1cc7d97278d91d6f332d2770233dd307" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "84933720ef4c9142d162818b36f46c78" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "cbcb186c78fcce62b0e60ad235ded077" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "79acd3ba35ef4bd7c05b0ff3619d25cc" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6275b9f008f2041cf6b55f9219935820" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "ed03750516a7150c38a56198f8ed15c8" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "9d641714998bd784d2b878277106dfc9" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "bf9f94ceb416af0bee1703b46c979daa" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "160ae492147e28a244464b104a57a579" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "46d85bc53cf41e4503a1dbb62a9b9ed8" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "a2c4089bb29bfe7d866d3d12a4c1032d" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d3639aec8dae74d41247e0a7e5610343" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0302c798ff13be981959e6768a2c6c33" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "55c8120f6934b03360bc0d41e4222993" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "dadaf90914794c4bbaa0f471ad9f2ffc" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "7bd535d3dbdd73ec099e0a7f763fbd54" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "15ca576ddc0f049fb369c38b05acdc2f" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "b7bd75e020541fb076aba0264428a8fe" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "65b83fe9e4df781cce1b84dd1093c87f" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "315de7c4fa927aa565dba4c3403ebee8" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "aca8dbd88876d32baff98ea7fc0b909b" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "fbba7234eadfe44138681c2aaf245cac" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "a3070ff22ea26d0b87c6d806a92eb819" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "1333dadb53639b23ab574ad649d2c03d" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "742a73a4bf1364efab18169a5e08d212" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "32df0bfa2f3ff624243a88100a009087" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "1f5ee892b0f2d36651f464f58496a345" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a472d8533e8047d3eccbd5751685dd5c" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "a3dc44a510b6cecaa0ac8fa7752d189e" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "bfa546adbf0b793544264307a6026a85" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "f86174e1467cbc13ca0addccda1152f7" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "19fe0a84ba3e047320561ba693276d87" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6ca6dac3c86b1674c1f1a761f76342b9" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 532480, + "records": [ + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8192 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 16384 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24576 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32768 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 40960 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 49152 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 57344 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 65536 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 73728 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 81920 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 90112 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 98304 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 106496 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 114688 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 122880 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 131072 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 139264 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 147456 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 155648 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 163840 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 172032 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 180224 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 188416 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 196608 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 204800 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 212992 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 221184 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 229376 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 237568 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 245760 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 253952 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 262144 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 270336 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 278528 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 286720 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 294912 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 303104 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 311296 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 319488 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 327680 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 335872 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 344064 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 352256 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 360448 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 368640 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 376832 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 385024 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 393216 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 401408 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 409600 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 417792 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 425984 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 434176 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 442368 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 450560 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 458752 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 466944 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 475136 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 483328 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 491520 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 499712 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 507904 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 516096 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 524288 + } + ], + "md5sum": "658b617cb31091c9cfe1afb32305810c" + } + ] +} \ No newline at end of file