diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,12575 @@ +{ + "metadata": { + "ParamSize": 805, + "ParamBytes": 39688355840.0, + "BitsPerParam": 4.500214977904823 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 525336576, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 128256, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 525336576, + "byteOffset": 0 + } + ], + "md5sum": "6f5e74599227c40ab689174bb88c2800" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 65667072, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 128256, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65667072, + "byteOffset": 0 + } + ], + "md5sum": "db99bc516623056f007ae98a3c75aa7b" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "f7a1783c591f23d770040417970dd701" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "83d0fc4b0122de280f1bdd7f81d0d27f" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "cbfcbdbe8776e1fad546b00e6c14e849" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1b1ded1aa9a3c68158ab490c93559850" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "1bc4223e36096eb1e951e429a5308025" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "fb802aefe3b565ca32a6c6c3923a8fb4" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d9a999d22268fa98cb2585d9ac989da2" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 29392896, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 5242880 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 9437184 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 24117248 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 24133632 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 24150016 + } + ], + "md5sum": "9cd3789aa0c451a24ef2ce0305b04605" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ebe6bef2e21ad1093e8ef53eb1e49553" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b4b02c3ef1ade8dc6b7420ef9d1239b0" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "11dd577951de782447324f5dab9f4ee9" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "bccf57a8801c76a43c8c8b6336b5e04a" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "b822be096d5f20825a7c2e0e2e188bb5" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6f1ce602693d73fa5f92cbe7b417f4f0" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4f46dca6d0a13181fbf3af081811e5d5" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6982920d3e5378844011a456accb176f" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 28344320, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 0 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 4194304 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 18874368 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 18890752 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 18907136 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 24150016 + } + ], + "md5sum": "40269c60231442ebd841abd3168afc1c" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9338379683217c18f23745c10d54014d" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "18155eb577269f746ecfaf1031f61fdc" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "269a5b7ba6f3bd474edd3ff87c5f89b6" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a7edad19d0a8c2283bfd095f18c628d2" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6bafa7580f0650ef5edbdaf29ccf1ca2" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "ee49814d64be0a111193659d3f228f95" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "5e590be2aa17281d22c11ec6e41e6f78" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "36fcf268da4fadbda7c8b3cacd782dfc" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "70b133837036cfbeac050a9836ea9773" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7050f6351e279ab91d343e469c6a0503" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "8bb9e86e24f45e0294864f5a05d2a932" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "eca8c413e82304f4d41e867216501af9" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "a7906018f796777c94de03e1235b0f94" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6898759b6a1b1f85a9e508bcaaf68c80" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "62d0a6ccec34c8780b1ee7f7eb314d03" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "043726272d9ce733c9200eeba31ee0e8" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a0676755ea49284b33b70929cc31e6fe" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "c2b42ed2efc1e9fe01334aa9b156948c" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "5d865ffbe521c767c62d2c36ad1f61dd" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "a9206e78d3f6c523a2b71512783d9441" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "4662e722826056fe45409389b60b4825" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "109ccb0bdae2988e95b93dbe23f7e9a4" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "5d031e38b68b694d824b00b61fb4a634" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "bd2a66faa2f7316fc55ab1689578b3d9" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "660b66f50e7ee4b516343b04f750cc24" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c0b051f1d778611d95eba6d8c01c3af0" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c7bf795af5985484c06d9ad25f2c1a97" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "678586bc757b0331520eacd2f8361ebf" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "0b9c672a2c2961504ecb6a03d8c856fb" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "41dc499bf0377a0c27894bbf30e55573" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "24656de65354137fb22ba600784afb9a" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "14f7fa64ebf149cfd05a74d1449ebde1" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "26f6225dc5776e8366a0b4b5499451c2" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "96c1efec5ac67c683e87202de4266385" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "42149964791fa82f30ea2cf2de9f37a9" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "8e83ab28aa70ffcc548a5aafa3e67dd2" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "5f8c96536c97c2b00822e33534d9ae35" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c52afabd9ea9c89c40794273b7286df3" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "87e3b26b743ae27df328d1e9598b5b78" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "ddb6ccadff4df9f4c32733c956adfd7a" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6ffbf00abfc4992f002537bdc1b84a92" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "cc0a14a2f404966e445a180f6630a50e" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "59c18644ffcc273f50f6893b5189a0e7" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "62c3986629ed42e2f5adf58089a44e03" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0cb9e7703ae4ae58414beec652fd371d" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7e721b417ce449144e55252df965c48b" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e09dd1dec069cf4c34264759e13f00c7" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "67e8324762900ac7fe4b3be7e769057c" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "11d221b431d4cea70f3fb81b6ee780cc" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c02388711bad5f125b06af6698b6d95f" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "7d7971ebbf688d3150418ab435715130" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "d07d627612f970d32773c8b368a61989" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "328fb7194a0b88a74ae4e28df09cba48" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "446ba782fa98b28b16846523edaa1b62" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "153b7c69cf37240e76fa47600e45fda0" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "edff7c1c0c1351c54269eae2918dc7d7" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "8ef5eba4bdb322692e9d3e4178d7a89d" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "591f72938c2ea91cbbad778f4176bb1c" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "519d27a8c39cb2431ab73e1ea77d9c33" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "6adc628f512fbd6f68b07377f005e288" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "7efdc64a30213e367e233bf506ee6063" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "036adb101c53b95ca8627856b273e2a1" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "a939bce1a0d886f9e80a2950f8fad9f4" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "ff39117f3820fbd27e6718e0a0975b08" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6e4915b90194dbaea2fec3dc66f9793b" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b52594bb32431274f23a204b04e4eb22" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "c3e6954240b67ab0ed617c045c1fd250" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "fb0d8ea8c8da25ebc3006e4984511830" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "e481bdc46a6493d2565ab1f011cf4d8b" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4d9a9c1442465d89ca176f95d58fe3b2" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "84bd008a623a81d5ca23d4a553c8c4eb" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "3ea970ad965783cb3f311393a4fd8c8c" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "41d002699fd4f1309de9fc9e33b43971" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6e1531b06e6c41c08164fe064d273f4d" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "542db9d7d9a5bb1be3e103d908d6a4e3" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c05d768fafbad54a0356d2fd3640c7f5" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "57d3808109ff6367faffcc9cb34b3687" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "91eb1dac5abffe045fc363f5142b4db2" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "41a0b008b39956c01dbaa0ac77f9c599" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "e3c3cfb8edec3f6f3247641abd7f4d23" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6fd6dcfef5cf44d452c6d1f0fa441cde" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3b78c75da01f7f7692a3a7221be32d3d" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6d2b2f16b30437147cdd3e2336747d31" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "1458aee96bfaa4cdcc36a44fea3fa466" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "0e6eb0386c79eb0d4f355017d51c3e17" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "cce72bb84a8f4c0da8e1dbf71624bd94" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "43a96d17ac8d32b3539025f618d558f0" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "465045ea20f3bbf19b2372aa1a62caa6" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "2403d7c40b03113acbd1871145ac53dc" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "1ecc6679bf77b85442cbf8a3359c2a39" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "be3354129adcb336174df28214232df5" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2e4bb7d2f4eafc359d470568840ef80c" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c71b3b29ee081da4fb88b45f8580ef55" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9337bdcf73e0cb13f060a01f8788174d" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d12f5b4fbaebe380171383ec0bd74b0f" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "0e24c8ce4eb494fcf990f7fe613cb6b4" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "15f7e6c47228f5036e7bee0e451156fa" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d7fcdf15818a2f29cef70da67bc76c8c" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ef1a543e211e4becabf65f802d25005e" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "16aa3deab37e17397334573fb6463eeb" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f40452819487b6eecbad18da87424781" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "a5a031b9a3f0a75b8383f1c5788d0f2d" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9750b689f2b01f0e596a40727d5fccf4" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "4593e4b7513f9a7140635bf6e267c10d" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c8cd522156618d37f658b222b04bfd03" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "329ef02e065eec0ed8b4c310b3e00f79" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "842b9f8d739a471030b1e6b6ce5776d7" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "55152bcff6321d77651f46e8acc065ac" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "5dd76a1651f5bc0e0e007a08ba04f994" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2c94300c9e9159fec8a6850108ec9bdb" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "a8da8c542e1935d7e6e8a983916ab1cc" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "549325365966648cdc2b0b37c400ee50" + }, + { + "dataPath": "params_shard_131.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "01a57268b2797a658f61ad1471a394aa" + }, + { + "dataPath": "params_shard_132.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "d564df6cd6e016459cef70979e317e1e" + }, + { + "dataPath": "params_shard_133.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "723e2fe37b4386489c12f6b1ffc1af76" + }, + { + "dataPath": "params_shard_134.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "b8b37aabf671f0d01a75eb7ef716e15a" + }, + { + "dataPath": "params_shard_135.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0f2d68d79e926c87adb6344283d7b7b4" + }, + { + "dataPath": "params_shard_136.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c7bff6a03f9c932d3d18c27617243d03" + }, + { + "dataPath": "params_shard_137.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d5a5c97ecfc5ad3b60a5578786c568d9" + }, + { + "dataPath": "params_shard_138.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "00d52ed7bc4ca0ffff858c9328a6eede" + }, + { + "dataPath": "params_shard_139.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "a7d73b43e5269ab5d8c0b67add849415" + }, + { + "dataPath": "params_shard_140.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "18ca5fae8405abff77c06663f08d7d42" + }, + { + "dataPath": "params_shard_141.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "e2e3d9e023fe7effe11cced2ed7550b6" + }, + { + "dataPath": "params_shard_142.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "88728e3e3d2647a8b3944a3b195eca3c" + }, + { + "dataPath": "params_shard_143.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4eaa2480172f911336a401c44401bde3" + }, + { + "dataPath": "params_shard_144.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "6f344b82c7780c1f9f9a2ee6e5d19513" + }, + { + "dataPath": "params_shard_145.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "4a4e599a6ed96f87235ba2261ec0ee4a" + }, + { + "dataPath": "params_shard_146.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2feafb1e4b7f874f74ac7980d3e470d9" + }, + { + "dataPath": "params_shard_147.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "75cf0f21ab0a7f0d837d2c6662922db4" + }, + { + "dataPath": "params_shard_148.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1dd07660663036e2f1d3b9f7eb7f4e4d" + }, + { + "dataPath": "params_shard_149.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "b17b79a00599fe80bb27ce10fcc1d034" + }, + { + "dataPath": "params_shard_150.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "55e117e515d4bf1973cd9a29b6e87a56" + }, + { + "dataPath": "params_shard_151.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "16fb56c4fb5ed2b04bd78f4804b52d78" + }, + { + "dataPath": "params_shard_152.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "3b9864c8837e9151d556d8a7a430578e" + }, + { + "dataPath": "params_shard_153.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "bbc6ccb16b2a3f60a7ba6b017060d803" + }, + { + "dataPath": "params_shard_154.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "960ff57c964843f088be52574949aa12" + }, + { + "dataPath": "params_shard_155.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ac5b211739a30c50a7a5c4e3f6b3207e" + }, + { + "dataPath": "params_shard_156.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "832bf22413e726a4242eb7b41bc4ae64" + }, + { + "dataPath": "params_shard_157.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "b4923764cce29c5d680d9ead610b8407" + }, + { + "dataPath": "params_shard_158.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c7bf7c27957a39eb012d8f0ba34937a1" + }, + { + "dataPath": "params_shard_159.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "707639b35f3dedf8d2d0cf4628e0e367" + }, + { + "dataPath": "params_shard_160.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9083d99dd53ff63fbb45c4210201f451" + }, + { + "dataPath": "params_shard_161.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "5ecc3310c774a8cd13cc9d10076328e8" + }, + { + "dataPath": "params_shard_162.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "e96dc2b6cfd17d133cb02eab8b0271bc" + }, + { + "dataPath": "params_shard_163.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "4d4764a646730b18172367df8eb28557" + }, + { + "dataPath": "params_shard_164.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "37adf8a46753913c3e913a51aab1a859" + }, + { + "dataPath": "params_shard_165.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "1697d4b55629717d75bcef3382939e92" + }, + { + "dataPath": "params_shard_166.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3054629bb410b61d23c4cd0d467b9e69" + }, + { + "dataPath": "params_shard_167.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "0bfdd76fc2ce834843bc1a57a9592b9b" + }, + { + "dataPath": "params_shard_168.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "66151a706cca0cfcf13ef3e182ebe4c5" + }, + { + "dataPath": "params_shard_169.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "7fc0fe5f099cfa12f136ff25b83d0882" + }, + { + "dataPath": "params_shard_170.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d3d19466530974d1ce877f3d2edb5ad0" + }, + { + "dataPath": "params_shard_171.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "a7e89855eec67b4a1e193f479a1e5b0e" + }, + { + "dataPath": "params_shard_172.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "85bae35685720078878a2378d51fef32" + }, + { + "dataPath": "params_shard_173.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "604297ea4f793df2fd675316a52cb908" + }, + { + "dataPath": "params_shard_174.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "76745db00e62886a70aa04f18528bc29" + }, + { + "dataPath": "params_shard_175.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "6b5d2acf2e8be2671bfb89db787ee9b7" + }, + { + "dataPath": "params_shard_176.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "bcb7f57170b1e63c194306ddde26bf39" + }, + { + "dataPath": "params_shard_177.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "87d79049d0a6be81a42e13cafd7beda2" + }, + { + "dataPath": "params_shard_178.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "315d82715d6ef1b7cd23312ce753b82e" + }, + { + "dataPath": "params_shard_179.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "255714be9e01d9463abbde03e5713c0b" + }, + { + "dataPath": "params_shard_180.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "a228e72ba9dd83031e48a13dd92d4281" + }, + { + "dataPath": "params_shard_181.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "0ad4d0c000d31e55c174d04f4e14461d" + }, + { + "dataPath": "params_shard_182.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "5b3c25707e873154b1f523e5117cce1b" + }, + { + "dataPath": "params_shard_183.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "a33b318d03f2902184f4ba0a340d2e10" + }, + { + "dataPath": "params_shard_184.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "48e756030a6b7b97547a24655a712eec" + }, + { + "dataPath": "params_shard_185.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "b3d2a6992933ba39c53f04d1b3477deb" + }, + { + "dataPath": "params_shard_186.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.30.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "e92e9f2c855851ce97de12c7e7357784" + }, + { + "dataPath": "params_shard_187.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "6c19b1f87a58e5e6cfd03fab5feb83af" + }, + { + "dataPath": "params_shard_188.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6ba430021cd6bf17eb85c8f3d942e70e" + }, + { + "dataPath": "params_shard_189.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "f6ecce4e1bc0c2d692465af9dcb1cd17" + }, + { + "dataPath": "params_shard_190.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f441d44d2286a121fe9a511d83edb366" + }, + { + "dataPath": "params_shard_191.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "579447addcf4a7ce16058ac1ed87a218" + }, + { + "dataPath": "params_shard_192.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.31.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "7257b70cfed1bc42116080579af1ba7f" + }, + { + "dataPath": "params_shard_193.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.32.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "2a4e62c519189f5b4fd2d5e9c69d6b89" + }, + { + "dataPath": "params_shard_194.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.32.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "be17e08702b11d801aa37448120642ad" + }, + { + "dataPath": "params_shard_195.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "9475a09dbb54549f2c31675596107758" + }, + { + "dataPath": "params_shard_196.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "617a16f3f332e1467fdc2368b86a8295" + }, + { + "dataPath": "params_shard_197.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "034d4a7ea29f196c81809f009b7bccc6" + }, + { + "dataPath": "params_shard_198.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.32.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.32.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "10f49b31ed8ffee563ee8d0b8736c5dd" + }, + { + "dataPath": "params_shard_199.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.33.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "a0fd4c3845e38db25e398a2063ce56f7" + }, + { + "dataPath": "params_shard_200.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.33.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "69dfcb2e0983e5650381b9ba5140bbd0" + }, + { + "dataPath": "params_shard_201.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "dec404dbcc16182ac9a5a37ea2cbf035" + }, + { + "dataPath": "params_shard_202.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "bc1fb22804c3bd111eb25839dc231154" + }, + { + "dataPath": "params_shard_203.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6de1e368c25d64f26d28151f482181db" + }, + { + "dataPath": "params_shard_204.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.33.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.33.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "eebab0464c9abcdd6b77c3cbefa7b42c" + }, + { + "dataPath": "params_shard_205.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.34.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "60ff484f88993e03a8f453badcfda390" + }, + { + "dataPath": "params_shard_206.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.34.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "03e7349beb373ca0e9e4deb206fc4b0f" + }, + { + "dataPath": "params_shard_207.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "cf1fc5e3fefc97b4a8ca3e7cee70c768" + }, + { + "dataPath": "params_shard_208.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "dffd13c3b184a64c39cf3acfaf5dc58c" + }, + { + "dataPath": "params_shard_209.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "cf53ca89d73b601a9f177b182d6092f2" + }, + { + "dataPath": "params_shard_210.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.34.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.34.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "1c49c190226fc8211712f3b1b8653920" + }, + { + "dataPath": "params_shard_211.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.35.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "22055b8b9aaa4595fbfa7df844b79762" + }, + { + "dataPath": "params_shard_212.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.35.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "ded93e0d85f45316060017716dc15dc4" + }, + { + "dataPath": "params_shard_213.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "72e75a828a24fe61b446d57651b5bce8" + }, + { + "dataPath": "params_shard_214.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "2992a932a146961b345a7079a9664452" + }, + { + "dataPath": "params_shard_215.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d177d163b6940fe6c0a61b9063ebb6f9" + }, + { + "dataPath": "params_shard_216.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.35.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.35.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "fb382ac4c04565529c3fc8828dfb6233" + }, + { + "dataPath": "params_shard_217.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.36.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "eaf5134b38bf2e24555b125503482e5d" + }, + { + "dataPath": "params_shard_218.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.36.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "35cd7f1d1cc41508ff6ec3582e6a91a6" + }, + { + "dataPath": "params_shard_219.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.36.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "1d090f84737bc1bd048cbba2c80fe5f6" + }, + { + "dataPath": "params_shard_220.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.36.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "efe9987c4094c7c62598b994f813e361" + }, + { + "dataPath": "params_shard_221.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "abff04e8aa47e69f0a41209476abeea3" + }, + { + "dataPath": "params_shard_222.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.36.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.36.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "94e1d1bd1f7dcc6509707afe6731d1c4" + }, + { + "dataPath": "params_shard_223.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.37.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9a2834319370e6dadc2efccc65edb752" + }, + { + "dataPath": "params_shard_224.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.37.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6017e008a73567c76fe946cfb36492ef" + }, + { + "dataPath": "params_shard_225.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.37.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "71ead886a3687e6ca0c1b2babd6116e5" + }, + { + "dataPath": "params_shard_226.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.37.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "424cb342ff21a85a6e9108f3603fdc51" + }, + { + "dataPath": "params_shard_227.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "3a4b6da277c8d572eb94590aeb50ba70" + }, + { + "dataPath": "params_shard_228.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.36.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.36.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.37.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.37.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "eb98afac840141f12510c9d1d4df6833" + }, + { + "dataPath": "params_shard_229.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.38.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "53cab22161f19f5ce3613a11dea94b50" + }, + { + "dataPath": "params_shard_230.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.38.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "a28706990b763ed0a5019593382ecf03" + }, + { + "dataPath": "params_shard_231.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "be2ebcc3c6b30fc06806a51333a90057" + }, + { + "dataPath": "params_shard_232.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "0ce9e87ccdb2a22b8d0ba97ba3c5ef2d" + }, + { + "dataPath": "params_shard_233.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.38.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f24069810d67104d77ef6d26a9a2e4d0" + }, + { + "dataPath": "params_shard_234.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.37.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.37.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.38.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.38.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "91f4c5717086fcca01d72fa14d06938d" + }, + { + "dataPath": "params_shard_235.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.39.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "6097e7c304dc09f3ddbf0c7502e542fa" + }, + { + "dataPath": "params_shard_236.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.39.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6fa3c0e2629c8a572c112cd2267b420c" + }, + { + "dataPath": "params_shard_237.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ff98c417af076bcae3648c142e58985a" + }, + { + "dataPath": "params_shard_238.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1bfcd469359cb955a933bbef2093ef95" + }, + { + "dataPath": "params_shard_239.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.39.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f181b9788adbf872e9ec6bc0876dfdfc" + }, + { + "dataPath": "params_shard_240.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.38.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.38.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.38.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.39.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.39.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "870ce34f5f103b7f1d0bfe3134f3d88f" + }, + { + "dataPath": "params_shard_241.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.40.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "489d2ad5a79666d6b9ea31d5bd95b18b" + }, + { + "dataPath": "params_shard_242.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.40.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "7d902127092aaed807918ffbae507bce" + }, + { + "dataPath": "params_shard_243.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.40.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "3f69ecf21255f098ee6aefaf5236e7b3" + }, + { + "dataPath": "params_shard_244.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.40.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "81239fbc9ef662ddfa1938f32624b5cf" + }, + { + "dataPath": "params_shard_245.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.40.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d9a806edd5bb4e79ec1150568fcbccd7" + }, + { + "dataPath": "params_shard_246.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.39.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.39.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.39.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.40.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.40.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "3d78984dc0a0a02e3eb9a9cd69bf82a9" + }, + { + "dataPath": "params_shard_247.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.41.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "26200b505cf2690fbed14c37d65821a2" + }, + { + "dataPath": "params_shard_248.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.41.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2621ad8bcf3c8a8498257c30b42099b2" + }, + { + "dataPath": "params_shard_249.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.41.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "34acf2118b6829c5b6de8f2b3ca10196" + }, + { + "dataPath": "params_shard_250.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.41.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "668badfe5a4f59b33156fde26c3aadb6" + }, + { + "dataPath": "params_shard_251.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "48787f742903d7e55b667953b999b6c1" + }, + { + "dataPath": "params_shard_252.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.40.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.40.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.40.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.41.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.41.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "8aec31e865b525b2e8ea0d3fe71b6685" + }, + { + "dataPath": "params_shard_253.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.42.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "1a8cb1e6eb17920abb866645d9c18dc3" + }, + { + "dataPath": "params_shard_254.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.42.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "31058e7c4c35deff580dba6ffd176709" + }, + { + "dataPath": "params_shard_255.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.42.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "7b7b8106ec5847f1e305b195ef98700b" + }, + { + "dataPath": "params_shard_256.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.42.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f9995ae3d7ab4b5247672f8864065128" + }, + { + "dataPath": "params_shard_257.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "56a29a8e09ad4e0c40c3e4e5d592935e" + }, + { + "dataPath": "params_shard_258.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.41.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.41.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.42.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.42.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "7df87c317ac9a49339e8fc1a5abe4658" + }, + { + "dataPath": "params_shard_259.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.43.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "6b7578bd18d7328550c4455a551512dc" + }, + { + "dataPath": "params_shard_260.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.43.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c3a680dacf8e6d842873473fe23f5f44" + }, + { + "dataPath": "params_shard_261.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.43.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d750d2c9478dc6ba38d1c1d0e2931718" + }, + { + "dataPath": "params_shard_262.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.43.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "299d107fd9ab3123adc1286fe2300586" + }, + { + "dataPath": "params_shard_263.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "b64c91f7c46f48746500e83c606ea38b" + }, + { + "dataPath": "params_shard_264.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.42.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.42.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.43.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.43.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b20f103c54407594a7fe20788749370f" + }, + { + "dataPath": "params_shard_265.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.44.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "466b481c6ba5deeb13ad957b9e9eb372" + }, + { + "dataPath": "params_shard_266.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.44.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d0ac3a48d455ed0df1ec76b87bd16bad" + }, + { + "dataPath": "params_shard_267.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.44.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "2ac2d16d059cfc3574e521610bc2927f" + }, + { + "dataPath": "params_shard_268.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.44.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "de663dfdf2d5945384ce2b67b148f154" + }, + { + "dataPath": "params_shard_269.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f36923ecfe7a5bd505ef8559d7899796" + }, + { + "dataPath": "params_shard_270.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.43.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.43.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.44.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.44.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "395c4a5552a59a618a7d626cce5d7208" + }, + { + "dataPath": "params_shard_271.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.45.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "4b6d7179a70785561a0a6ed5309888ff" + }, + { + "dataPath": "params_shard_272.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.45.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c679a0df3f5e92e6d3339b8ac428e796" + }, + { + "dataPath": "params_shard_273.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.45.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "e1b81c070dbeceb39135d4c202af362b" + }, + { + "dataPath": "params_shard_274.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.45.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "d3d8d8352e0830c282af512cbf3a7d2a" + }, + { + "dataPath": "params_shard_275.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "091adae1147662150af5863f532cb72e" + }, + { + "dataPath": "params_shard_276.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.44.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.44.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.45.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.45.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "396c4f3dc36e328a599375c5ada2f318" + }, + { + "dataPath": "params_shard_277.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.46.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "715910e287747b7f7d20d03954f34b6d" + }, + { + "dataPath": "params_shard_278.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.46.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c9b6b2aa642ce220c5918318d5c45769" + }, + { + "dataPath": "params_shard_279.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.46.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "cf7a834f9ec146d7a956b24241e80367" + }, + { + "dataPath": "params_shard_280.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.46.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "fef0150714567dcdaeed6f925a84d1b7" + }, + { + "dataPath": "params_shard_281.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "19f7a50e653239f72ffd1c929882baeb" + }, + { + "dataPath": "params_shard_282.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.45.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.45.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.46.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.46.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "310010d9ee4c241269a85f524502ebdb" + }, + { + "dataPath": "params_shard_283.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.47.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "7dcf3ea58586cbb536c537e6454b5524" + }, + { + "dataPath": "params_shard_284.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.47.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "6e0e702c56ee82481f2820102e8a0a69" + }, + { + "dataPath": "params_shard_285.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.47.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "68c353415dac3c064d86dbc5b3a16960" + }, + { + "dataPath": "params_shard_286.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.47.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "14e025995405cceeb242ad7a245c276f" + }, + { + "dataPath": "params_shard_287.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "0facefef3618b7a601e7210755363bed" + }, + { + "dataPath": "params_shard_288.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.46.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.46.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.47.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.47.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "2e7c4727372e41e64b66c0724f5d667e" + }, + { + "dataPath": "params_shard_289.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.48.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "4797897507e5d90a44718d86a5d7411b" + }, + { + "dataPath": "params_shard_290.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.48.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "eb6726bdc774170d5225df58501f1668" + }, + { + "dataPath": "params_shard_291.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.48.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d78180e57b1cd29a82421939521988a6" + }, + { + "dataPath": "params_shard_292.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.48.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b86236bee5046b03321d60c6538d813b" + }, + { + "dataPath": "params_shard_293.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.48.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f016dafc758f955c246e93ece14fe0cb" + }, + { + "dataPath": "params_shard_294.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.47.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.47.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.48.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.48.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "1c6c14f1c31e217f16513410ce74b34e" + }, + { + "dataPath": "params_shard_295.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.49.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "8c8e7b46befbc184de00d71f0d7b49e5" + }, + { + "dataPath": "params_shard_296.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.49.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "23e4c1efdc807e598c5653a7f59ae8b6" + }, + { + "dataPath": "params_shard_297.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.49.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "13ef6b657a238ec4159ca0c9ea4237c9" + }, + { + "dataPath": "params_shard_298.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.49.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "85ef4c5b54532dd908474a52a0d48807" + }, + { + "dataPath": "params_shard_299.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.49.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "668977a5e9c9f463946548009888bda1" + }, + { + "dataPath": "params_shard_300.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.48.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.48.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.48.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.49.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.49.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "4126e62458335028c9409c1e68b455f5" + }, + { + "dataPath": "params_shard_301.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.50.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "b1362ed474575b0ad77ad974e09e84d1" + }, + { + "dataPath": "params_shard_302.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.50.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "feda34586438bd54523d40ccf1102799" + }, + { + "dataPath": "params_shard_303.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.50.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "addd223cc24ef648a37bf79846e1a113" + }, + { + "dataPath": "params_shard_304.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.50.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a26e41323ddcb083a6dcbdead33080b1" + }, + { + "dataPath": "params_shard_305.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.50.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "b2caa9261542dbda273e0a667834b6ba" + }, + { + "dataPath": "params_shard_306.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.49.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.49.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.49.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.50.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.50.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "3ffaf477c1ae8230721ab326e606c72c" + }, + { + "dataPath": "params_shard_307.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.51.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "a3b71316acdccadcbe03e64adef43f7a" + }, + { + "dataPath": "params_shard_308.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.51.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "69e8430806fe7138ec9463e9590f44a8" + }, + { + "dataPath": "params_shard_309.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.51.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6e0d7e744c6c162521be704168488700" + }, + { + "dataPath": "params_shard_310.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.51.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "6c747e10c677c6c7d9046f1b72688403" + }, + { + "dataPath": "params_shard_311.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.51.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d24915e43cfa356b61221313e5c2b58b" + }, + { + "dataPath": "params_shard_312.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.50.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.50.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.50.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.51.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.51.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b0fb953a904c0e92c57807d5b260d32a" + }, + { + "dataPath": "params_shard_313.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.52.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "b66b297d0ab1ff2e34221ec79a1198f5" + }, + { + "dataPath": "params_shard_314.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.52.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2d8bd855a487d31e2275f4e282aedd27" + }, + { + "dataPath": "params_shard_315.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.52.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "f1a30af02430f98aef6c12577eadb9cf" + }, + { + "dataPath": "params_shard_316.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.52.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "46058db992963bbab0b78d6e890b6b92" + }, + { + "dataPath": "params_shard_317.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.52.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ebfb365e591b36e00edf31303701f936" + }, + { + "dataPath": "params_shard_318.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.51.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.51.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.51.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.52.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.52.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "ed84db204666941f82e57da6d2a20a79" + }, + { + "dataPath": "params_shard_319.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.53.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "8806bfb23abea78c84232e20babfc9bd" + }, + { + "dataPath": "params_shard_320.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.53.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "8c2335087d48a6a2016bdebed1c59ae7" + }, + { + "dataPath": "params_shard_321.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.53.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ffd7d595880a237148cd8ae355ba602a" + }, + { + "dataPath": "params_shard_322.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.53.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "55e206309e7a35ab7b5f66ad63c60bec" + }, + { + "dataPath": "params_shard_323.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.53.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "5dc99f9d4917b6fd2a1c90dc3b09da51" + }, + { + "dataPath": "params_shard_324.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.52.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.52.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.52.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.53.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.53.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "855bb5a566e41bfeeb834be6f81dc405" + }, + { + "dataPath": "params_shard_325.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.54.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9c56f32df3b7d481acf24dac25df7972" + }, + { + "dataPath": "params_shard_326.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.54.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "ef4ed9183f752eed57ac2428447852e8" + }, + { + "dataPath": "params_shard_327.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.54.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "85186906e9571253ad27edb540b06ec7" + }, + { + "dataPath": "params_shard_328.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.54.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "68124a12cfa492a9c7a5506b0e2bcb18" + }, + { + "dataPath": "params_shard_329.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.54.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "975e330ea6712b583d66f68bee36c7b8" + }, + { + "dataPath": "params_shard_330.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.53.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.53.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.53.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.54.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.54.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "15ea420a140c9d98f158f07fc57c1b03" + }, + { + "dataPath": "params_shard_331.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.55.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "d15415c9bb221f4e031ca192c61c1a69" + }, + { + "dataPath": "params_shard_332.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.55.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "1a288f49ad800f6637bfdb640148c687" + }, + { + "dataPath": "params_shard_333.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.55.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "514e9b09765dd88b5ff2dda90ac2a70b" + }, + { + "dataPath": "params_shard_334.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.55.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4b0856512de0909264383098551a1bdf" + }, + { + "dataPath": "params_shard_335.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.55.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e5a14db12b187563022f03232172b310" + }, + { + "dataPath": "params_shard_336.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.54.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.54.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.54.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.55.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.55.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "c982e6434ab28b3bf6d45af088802057" + }, + { + "dataPath": "params_shard_337.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.56.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "1bfd45bc7b133d34b11e7fdbaf369b74" + }, + { + "dataPath": "params_shard_338.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.56.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "e6691e95231d873a45e1631cd0fa64db" + }, + { + "dataPath": "params_shard_339.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.56.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "273c5f87c35cbfc5f908e1c0a4bf0e82" + }, + { + "dataPath": "params_shard_340.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.56.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "8f61b31cef41b820e26d11d5ae1e6747" + }, + { + "dataPath": "params_shard_341.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.56.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "15129acdd1cfa244a40f7b39219e40ab" + }, + { + "dataPath": "params_shard_342.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.55.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.55.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.55.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.56.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.56.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "37652e3ba6f2241f9145940a2b0bbc8a" + }, + { + "dataPath": "params_shard_343.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.57.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "30e53772be7b43b485ab88a509fb1332" + }, + { + "dataPath": "params_shard_344.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.57.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "7b054392dbd00e565e5a4f4a57915468" + }, + { + "dataPath": "params_shard_345.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.57.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ce8cb8f6b86e0739e116f8abe35f620e" + }, + { + "dataPath": "params_shard_346.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.57.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "29e41e33adfffc0f23ffa5a6bf382798" + }, + { + "dataPath": "params_shard_347.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.57.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "8b761fc8df907c575b42cd3a5db17fa9" + }, + { + "dataPath": "params_shard_348.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.56.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.56.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.56.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.57.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.57.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "248e1a6ab692d4f05d53437469bef209" + }, + { + "dataPath": "params_shard_349.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.58.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "28af4885dbe406e5554ec582605d067e" + }, + { + "dataPath": "params_shard_350.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.58.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c65863a797fcd26973670cf93ca1a37a" + }, + { + "dataPath": "params_shard_351.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.58.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "5d81c267914b4ad517cae714910f7359" + }, + { + "dataPath": "params_shard_352.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.58.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "22650bbc1c69fa27dddae1c74f4e8e47" + }, + { + "dataPath": "params_shard_353.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.58.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "267b00fefe99b5563457a4730e189691" + }, + { + "dataPath": "params_shard_354.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.57.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.57.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.57.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.58.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.58.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "f1468e3496cee916cf79079c2deedf51" + }, + { + "dataPath": "params_shard_355.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.59.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "3c508649688c0d1c29a3990bdf63fdbd" + }, + { + "dataPath": "params_shard_356.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.59.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "167dabe73b1deba4eb06ef53b4d6db8f" + }, + { + "dataPath": "params_shard_357.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.59.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "99e214ef736d822c0d45f6967a387b7d" + }, + { + "dataPath": "params_shard_358.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.59.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "0882637a5a47f2fc8d895b97e19ca1f2" + }, + { + "dataPath": "params_shard_359.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.59.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ef763d04b7984523dbdfb8d7cc321a6f" + }, + { + "dataPath": "params_shard_360.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.58.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.58.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.58.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.59.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.59.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b692b3aabd00dfbd7a2c4526eca0829c" + }, + { + "dataPath": "params_shard_361.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.60.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "3b9c7a8a1b62f97805ff022200fa5dde" + }, + { + "dataPath": "params_shard_362.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.60.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "7120d34e9fa01b7ca77c2292c3469558" + }, + { + "dataPath": "params_shard_363.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.60.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "e7f5e2acf1208413b770a1726c56254e" + }, + { + "dataPath": "params_shard_364.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.60.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "765964437e9b03a2ae951f3bf70eb9af" + }, + { + "dataPath": "params_shard_365.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.60.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6f576b1c6a4f0ee9e153cb46f0dc180f" + }, + { + "dataPath": "params_shard_366.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.59.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.59.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.59.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.60.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.60.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "e9fb54e0e15bb81ee0719eaf74d448d4" + }, + { + "dataPath": "params_shard_367.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.61.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "edaaa55343007b7a32a8e0981b122f7c" + }, + { + "dataPath": "params_shard_368.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.61.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "f160724dd200ce47f7fbf0d733828ada" + }, + { + "dataPath": "params_shard_369.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.61.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "477c1fd7af280335c11cafd68d81556d" + }, + { + "dataPath": "params_shard_370.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.61.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "25c71f7f3b9d48394f0f930204a56b23" + }, + { + "dataPath": "params_shard_371.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.61.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "2774abe21d787f9020c9839a4faf565f" + }, + { + "dataPath": "params_shard_372.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.60.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.60.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.60.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.61.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.61.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "d9cf309f14543beb9c54ee758d20066f" + }, + { + "dataPath": "params_shard_373.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.62.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "6dbe1cf8f0a476ea64485640be12e909" + }, + { + "dataPath": "params_shard_374.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.62.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "94e1b87a01381e807f6ac5dedcb82fea" + }, + { + "dataPath": "params_shard_375.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.62.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "5b9eb5c37c6ea03d00b433aec7d8ab25" + }, + { + "dataPath": "params_shard_376.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.62.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a97e57b1a2c112096d5fdd92c35587c5" + }, + { + "dataPath": "params_shard_377.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.62.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "98fc4fdc3aa28ae8ad53fba8d471f66b" + }, + { + "dataPath": "params_shard_378.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.61.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.61.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.61.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.62.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.62.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "83c7888016389785ca8e9466b2b41671" + }, + { + "dataPath": "params_shard_379.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.63.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "8f817272f3eeabe5e1a7d2c7ca4813b9" + }, + { + "dataPath": "params_shard_380.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.63.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "82531ebc2e4ac14233dfb329015f69a1" + }, + { + "dataPath": "params_shard_381.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.63.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "4a24278378e2abc5125803b2ff25bf8c" + }, + { + "dataPath": "params_shard_382.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.63.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4c95ec67ed15f612d33813c02685e16f" + }, + { + "dataPath": "params_shard_383.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.63.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "0d43a30afa7f57c53b679a80a6616be5" + }, + { + "dataPath": "params_shard_384.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.62.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.62.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.62.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.63.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.63.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b7a3b05f7fe1a3962f3b78163a982285" + }, + { + "dataPath": "params_shard_385.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.64.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "5ec043c40e16d66c2e8ffa26be952c5b" + }, + { + "dataPath": "params_shard_386.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.64.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "b7ef21a90910a12803d4e79f50be3bad" + }, + { + "dataPath": "params_shard_387.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.64.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "0a3a968c7ca9780843ac0766e520ddad" + }, + { + "dataPath": "params_shard_388.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.64.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c7598f52110b0c1d840d9431773fd10f" + }, + { + "dataPath": "params_shard_389.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.64.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a0f31d8012bc48d6ee15ce183afa70eb" + }, + { + "dataPath": "params_shard_390.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.63.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.63.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.63.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.64.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.64.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "f2ee3797604d8f4413c4df86c037c88d" + }, + { + "dataPath": "params_shard_391.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.65.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "f49c3e2d02b0ed8bb45adb2f4b11e650" + }, + { + "dataPath": "params_shard_392.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.65.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "4959a7a4fe61945e888ea3b8053e034d" + }, + { + "dataPath": "params_shard_393.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.65.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "809bca1ed0e570af03b5b96c9185b612" + }, + { + "dataPath": "params_shard_394.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.65.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "08ba42b52f1d902ac56eb765586052b0" + }, + { + "dataPath": "params_shard_395.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.65.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "2ec6752553b71385583b0cc85e13e3c8" + }, + { + "dataPath": "params_shard_396.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.64.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.64.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.64.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.65.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.65.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "47df40bf42fcbbe9330de8452db32c30" + }, + { + "dataPath": "params_shard_397.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.66.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "2065f2f7b07ccef9bc0b0542d92cccf9" + }, + { + "dataPath": "params_shard_398.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.66.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "642a43b9f74d436e71fbf453bba9d69e" + }, + { + "dataPath": "params_shard_399.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.66.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "85b20ed9697e240823af10e4868b648b" + }, + { + "dataPath": "params_shard_400.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.66.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c0fa16ad2c1cd8465ca27fb1e51a47bd" + }, + { + "dataPath": "params_shard_401.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.66.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9a33229ac8cb774e46a00b88d9529371" + }, + { + "dataPath": "params_shard_402.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.65.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.65.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.65.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.66.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.66.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "fae8cd60d019708966ff5e80cfa64af8" + }, + { + "dataPath": "params_shard_403.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.67.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "83e9753da1e94f333c935f2910fef1fe" + }, + { + "dataPath": "params_shard_404.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.67.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2f6ec29a28711373929ee90e3aed37c5" + }, + { + "dataPath": "params_shard_405.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.67.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "629fa50befe958db76d7bf6cb28466df" + }, + { + "dataPath": "params_shard_406.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.67.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "049b8a023c70714dc6474da294d8f872" + }, + { + "dataPath": "params_shard_407.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.67.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ad215d593ebaa2da71c478b9d7149bd7" + }, + { + "dataPath": "params_shard_408.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.66.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.66.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.66.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.67.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.67.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "76e2b53854bec34fc6320f039bc71ee1" + }, + { + "dataPath": "params_shard_409.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.68.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "717264a93e96af5b10f77b6a6f7d8737" + }, + { + "dataPath": "params_shard_410.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.68.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "344c6d2a06e3f13c06e6ebc993ba037d" + }, + { + "dataPath": "params_shard_411.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.68.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "fad70137074baf855af688f71c040c27" + }, + { + "dataPath": "params_shard_412.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.68.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "80b1e0bf9088a606d98c6ec653472b33" + }, + { + "dataPath": "params_shard_413.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.68.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f89dca7b82e8b5069b66dabe73073dd3" + }, + { + "dataPath": "params_shard_414.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.67.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.67.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.67.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.68.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.68.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "7f590d0262d8ba875a27cb701a031f30" + }, + { + "dataPath": "params_shard_415.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.69.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "bf1ffaac595e7a8ed91758567cf79ebe" + }, + { + "dataPath": "params_shard_416.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.69.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "7b0f78febdd88296b06ebf2711a578a9" + }, + { + "dataPath": "params_shard_417.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.69.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ddf82d110588d47cdc22df131fa807ac" + }, + { + "dataPath": "params_shard_418.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.69.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "222225c7430979e226c03e0ea62ebe01" + }, + { + "dataPath": "params_shard_419.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.69.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "3b42ef1fac67793c69aa827b39397b35" + }, + { + "dataPath": "params_shard_420.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.68.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.68.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.68.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.69.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.69.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "99995bdbef7d54ea786da0d5b832d74f" + }, + { + "dataPath": "params_shard_421.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.70.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9d5f0ef990487fd25bed24aac360171b" + }, + { + "dataPath": "params_shard_422.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.70.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "20948c4d080275a144c5a24c61a0c92a" + }, + { + "dataPath": "params_shard_423.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.70.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "b2ea5000615ee9682c82a7872bd1f8df" + }, + { + "dataPath": "params_shard_424.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.70.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1c68cb8832efc783671fb8f098cb7fe5" + }, + { + "dataPath": "params_shard_425.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.70.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a4da646f71d01ead17842f1ffacddd45" + }, + { + "dataPath": "params_shard_426.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.69.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.69.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.69.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.70.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.70.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "c0a6f2893b045c33c22f151618a06253" + }, + { + "dataPath": "params_shard_427.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.71.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "c2e2b6a220717684388cf37ed741106f" + }, + { + "dataPath": "params_shard_428.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.71.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "cac5f4f50f81b87092546d3707441d34" + }, + { + "dataPath": "params_shard_429.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.71.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "6b73ce62871d2d5746e15dacbcfca325" + }, + { + "dataPath": "params_shard_430.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.71.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "0c51867dea35d8deb532e6788ad49862" + }, + { + "dataPath": "params_shard_431.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.71.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "5a17824c2cf42b320ef0dafa8d8b3994" + }, + { + "dataPath": "params_shard_432.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.70.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.70.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.70.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.71.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.71.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "8416e348687dbf11348bab7768d07142" + }, + { + "dataPath": "params_shard_433.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.72.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "1fe2f4551503214952f2a863c70cdb3f" + }, + { + "dataPath": "params_shard_434.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.72.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "9bf8e01f1929549ec8174982124c42ad" + }, + { + "dataPath": "params_shard_435.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.72.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "8d2d1237f08242d5b7edd3a25bd0a392" + }, + { + "dataPath": "params_shard_436.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.72.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "02a79c6f1b017f2ac230f9f3759849e8" + }, + { + "dataPath": "params_shard_437.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.72.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "2ed8e8fa5dbe605671f9a14688bef3dc" + }, + { + "dataPath": "params_shard_438.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.71.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.71.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.71.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.72.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.72.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "5200be8c36edfe3091dc599caa9f2ec7" + }, + { + "dataPath": "params_shard_439.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.73.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "e61933faf96a5039317f2c75cd759e25" + }, + { + "dataPath": "params_shard_440.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.73.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "991aea932ad579c8bc7d05f1afa7d3ea" + }, + { + "dataPath": "params_shard_441.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.73.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "76cc1db24e45824a3c308abe7388285c" + }, + { + "dataPath": "params_shard_442.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.73.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b5e4c6333d8e1cd8b11c9686097eb10d" + }, + { + "dataPath": "params_shard_443.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.73.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "553471df3c6ee74dbeee63634c98da76" + }, + { + "dataPath": "params_shard_444.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.72.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.72.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.72.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.73.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.73.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "9af943f9256df0ef6e2f8a9e8229ec75" + }, + { + "dataPath": "params_shard_445.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.74.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "4b2b2e5af96b0bbbc7d6cffbbb9921aa" + }, + { + "dataPath": "params_shard_446.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.74.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "a1da62b4103b0a25a9fea785dbbd8ab4" + }, + { + "dataPath": "params_shard_447.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.74.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d8d5331cbda0cf3a477a04dbde6b2230" + }, + { + "dataPath": "params_shard_448.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.74.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "8a5b9e0f40e3c5b96adc8e5a4d9bd7b4" + }, + { + "dataPath": "params_shard_449.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.74.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "cbcb6b58f0855999704a4ecd9e113fde" + }, + { + "dataPath": "params_shard_450.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.73.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.73.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.73.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.74.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.74.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "9b2e87adb075eaae329d79238976c629" + }, + { + "dataPath": "params_shard_451.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.75.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "acf5d871253ab601d364684891f98f82" + }, + { + "dataPath": "params_shard_452.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.75.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "998fb08572d980ce0053b779ac546e2c" + }, + { + "dataPath": "params_shard_453.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.75.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "11fd66edd06539e66af314e1ec9af965" + }, + { + "dataPath": "params_shard_454.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.75.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a4260015fe3acf4a47d1f5bdfdc522ec" + }, + { + "dataPath": "params_shard_455.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.75.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9d67aa2513051d4862c0404fffd395d8" + }, + { + "dataPath": "params_shard_456.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.74.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.74.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.74.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.75.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.75.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "db6dc115490c934229e43a90553031ce" + }, + { + "dataPath": "params_shard_457.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.76.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "9bdc1c435b9f50bc4c30935439afb653" + }, + { + "dataPath": "params_shard_458.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.76.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "00110c365e91fdf6895eeb679e929002" + }, + { + "dataPath": "params_shard_459.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.76.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "7098213bc764ad70fdc5a57c385daac8" + }, + { + "dataPath": "params_shard_460.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.76.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "e249634614a836bed13684dbe61a6c3e" + }, + { + "dataPath": "params_shard_461.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.76.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "fb4d4c324e19709dc8a1a1f04dd2b736" + }, + { + "dataPath": "params_shard_462.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.75.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.75.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.75.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.76.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.76.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "243237052f648c5a163b970ad323a269" + }, + { + "dataPath": "params_shard_463.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.77.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "df5360a730ad561471839489b095482d" + }, + { + "dataPath": "params_shard_464.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.77.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "99c3af8be2da0d91803fd539d0219f35" + }, + { + "dataPath": "params_shard_465.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.77.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "125ca1f13ae67a151c6bf8c401524f2c" + }, + { + "dataPath": "params_shard_466.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.77.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3eda4b34fc1685930925fdbd4d965de3" + }, + { + "dataPath": "params_shard_467.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.77.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "206d7806862e19e0e7233c393e030615" + }, + { + "dataPath": "params_shard_468.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.76.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.76.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.76.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.77.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.77.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "b8e1937be61c1aa0186e4d6899f920b6" + }, + { + "dataPath": "params_shard_469.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.78.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "fa1401a75b7d9026574cee2a835346f5" + }, + { + "dataPath": "params_shard_470.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.78.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "90b054629563d4ac9413d864a640bcf1" + }, + { + "dataPath": "params_shard_471.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.78.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "413700dae0cbe4bbcf612907bfad3095" + }, + { + "dataPath": "params_shard_472.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.78.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1a78366683615c50791f90c53eb448d3" + }, + { + "dataPath": "params_shard_473.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.78.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ccc664e9ce8c644439849f2de8f98015" + }, + { + "dataPath": "params_shard_474.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.77.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.77.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.77.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.78.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.78.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "dabd93d7ba32d09c4ae8bb2ad1ca4b7f" + }, + { + "dataPath": "params_shard_475.bin", + "format": "raw-shard", + "nbytes": 41943040, + "records": [ + { + "name": "model.layers.79.self_attn.qkv_proj.q_weight", + "shape": [ + 10240, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 41943040, + "byteOffset": 0 + } + ], + "md5sum": "e45f9a271bf1c96b5eaf6b13d503f3d7" + }, + { + "dataPath": "params_shard_476.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.79.self_attn.o_proj.q_weight", + "shape": [ + 8192, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "0a01fd6615c225eb55791bef63990e92" + }, + { + "dataPath": "params_shard_477.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.79.mlp.gate_up_proj.q_weight", + "shape": [ + 57344, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "baf19b48fa97bd7f282a0563a9d754e1" + }, + { + "dataPath": "params_shard_478.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.79.mlp.gate_up_proj.q_scale", + "shape": [ + 57344, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "980a76eac98b438eaeb9fe04903d37db" + }, + { + "dataPath": "params_shard_479.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.79.mlp.down_proj.q_weight", + "shape": [ + 8192, + 3584 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "26c189280f5b1594caae139a3c2c7b54" + }, + { + "dataPath": "params_shard_480.bin", + "format": "raw-shard", + "nbytes": 24150016, + "records": [ + { + "name": "model.layers.78.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.78.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.78.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.layers.79.self_attn.qkv_proj.q_scale", + "shape": [ + 10240, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 5242880, + "byteOffset": 14712832 + }, + { + "name": "model.layers.79.self_attn.o_proj.q_scale", + "shape": [ + 8192, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4194304, + "byteOffset": 19955712 + } + ], + "md5sum": "c059f9ebdf234dbb1630b8f4f271e633" + }, + { + "dataPath": "params_shard_481.bin", + "format": "raw-shard", + "nbytes": 525336576, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 128256, + 1024 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 525336576, + "byteOffset": 0 + } + ], + "md5sum": "730787cac5ba05aea86d7cb9a73cc13c" + }, + { + "dataPath": "params_shard_482.bin", + "format": "raw-shard", + "nbytes": 65667072, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 128256, + 256 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65667072, + "byteOffset": 0 + } + ], + "md5sum": "49eb95dcae5fdec4009c154ea486cd4b" + }, + { + "dataPath": "params_shard_483.bin", + "format": "raw-shard", + "nbytes": 14729216, + "records": [ + { + "name": "model.layers.79.mlp.down_proj.q_scale", + "shape": [ + 8192, + 896 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14680064, + "byteOffset": 0 + }, + { + "name": "model.layers.79.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14680064 + }, + { + "name": "model.layers.79.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14696448 + }, + { + "name": "model.norm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 14712832 + } + ], + "md5sum": "4bc634d100b8d489d80dacab3038d990" + } + ] +} \ No newline at end of file