{ "metadata": { "ParamSize": 398, "ParamBytes": 9571174400.0, "BitsPerParam": 5.223190258401046 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 1027604480, "records": [ { "name": "lm_head.weight", "shape": [ 100352, 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1027604480, "byteOffset": 0 } ], "md5sum": "20de31aeb1ef1b99e6cf8c806f36a376" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.34.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "66861053e8f9db1bba3e6fdbcccebe53" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.35.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "bde6752771741e9481caaeec3598933d" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.35.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "069d93efdb0fd34611b361cda1f833ae" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 26050560, "records": [ { "name": "model.layers.34.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 0 }, { "name": "model.layers.34.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 10240 }, { "name": "model.layers.34.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 1443840 }, { "name": "model.layers.35.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 1454080 }, { "name": "model.layers.35.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 1464320 }, { "name": "model.layers.35.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 2897920 }, { "name": "model.layers.35.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5765120 }, { "name": "model.layers.35.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 5775360 }, { "name": "model.layers.35.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 25436160 } ], "md5sum": "6d988e1a7242b7709dcd98e9382cb1ce" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.36.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "eb11771a13aaa88339be5adccade49ae" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.36.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "9b78167bf96252fa1058150ebf1a5a12" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.36.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "2eadf4e4e42a52196573c86cd094346e" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.37.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "824c868681cd4bbcf3bf689acb425e6c" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.37.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "f8f7b26d523899de3f25d067d4657757" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.35.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.35.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.36.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.36.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.36.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.36.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.36.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.36.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.36.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.37.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.37.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "3727a999c25839d1b7f5f49e32514987" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 52428800, "records": [ { "name": "model.layers.37.self_attn.o_proj.weight", "shape": [ 5120, 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 52428800, "byteOffset": 0 } ], "md5sum": "23764fff9750e6c52a046f14df36c478" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.38.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "00d9a77f40857359c42d3fd692768811" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.38.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "4651f2a8429a53226d062e8bd9837e00" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.38.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "2727640e5459ce0a217e801dce9122e5" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 28088320, "records": [ { "name": "model.layers.37.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.37.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.37.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.37.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 }, { "name": "model.layers.38.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 23152640 }, { "name": "model.layers.38.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 23162880 }, { "name": "model.layers.38.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 24596480 }, { "name": "model.layers.38.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27463680 }, { "name": "model.layers.38.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 27473920 } ], "md5sum": "38df28635ee15b284bc902e7bfe3e147" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.39.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "2c6ab3305a171954c69afaf314aa3a21" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.39.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "f465df2198b516f7800df891444e36bb" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.39.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "d4b732c2fc2982a57c7bd46fecab6cbd" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 52428800, "records": [ { "name": "model.layers.39.self_attn.o_proj.weight", "shape": [ 5120, 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 52428800, "byteOffset": 0 } ], "md5sum": "3a865be7fc173ed9cc8b397291dcca07" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 1027604480, "records": [ { "name": "model.embed_tokens.weight", "shape": [ 100352, 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1027604480, "byteOffset": 0 } ], "md5sum": "98c1a2ba2fb4ab48ad97bd78c54eed1d" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "5a9864f2cfb172fc37228a5839d863a1" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "6f79c27eee60b6a9ac4045a327a4a92d" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "36ac259109cb3067210790d0d903cf99" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 23398400, "records": [ { "name": "model.layers.38.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.38.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.39.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.39.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.39.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.39.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.39.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.norm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18452480 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18462720 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 18472960 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 19906560 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 22773760 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22784000 } ], "md5sum": "e51c62c0571eedb8b850b308eedf27ac" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 183500800, "records": [ { "name": "model.layers.1.mlp.down_proj.weight", "shape": [ 5120, 17920 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 183500800, "byteOffset": 0 } ], "md5sum": "a8fffb756d1e96786accb77a23f96813" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "8cf0a1025dce43bebcca58efd8068da9" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "d66ba4997c01f7c865d7374e1a968f4c" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 183500800, "records": [ { "name": "model.layers.2.mlp.down_proj.weight", "shape": [ 5120, 17920 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 183500800, "byteOffset": 0 } ], "md5sum": "09346a746b1441926cbbc0c88ade1a19" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "abee93723ef8f5a8dd00a8007efd59a4" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "c5e892a72f6150e904335eb26312fc63" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 33423360, "records": [ { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 13527040 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 16394240 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 16404480 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 17018880 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 30126080 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 30535680 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 30545920 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 33413120 } ], "md5sum": "2c2a2fdf5118411a4681fc20126b5cb1" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 183500800, "records": [ { "name": "model.layers.3.mlp.down_proj.weight", "shape": [ 5120, 17920 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 183500800, "byteOffset": 0 } ], "md5sum": "b9f36fb68508fc3764ddceacedd88dcb" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "8960cdf13818af64a3fa1a187855f3f0" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "42a597c6ac8d9f8bdcff4acadbd6a7ab" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "44c92b9d1d2b7d90a421f46e2d69ad71" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "7c2edc364bac851447b01a985bed6617" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 32593920, "records": [ { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 0 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 614400 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13721600 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 14131200 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14141440 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17008640 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17018880 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 17633280 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 30740480 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31150080 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31160320 } ], "md5sum": "a3fe31f7b9912607d59f292f4eb48184" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "859d59ea84c539c843476380d941ad26" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "e5e234e3aed92bc5fabac075ed68e328" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "7a684e5fafa8cbc8ca35da1abbe14aca" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "dd9074ec115dbe302e82544e3cf00d2f" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "d80136d9714092a69380505a236128d8" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 31959040, "records": [ { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 13516800 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 16384000 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 16998400 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 30105600 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 30515200 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 30525440 } ], "md5sum": "c0fe8e856a1d6c4dbceba3de803e6483" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "3b9bedb5fd1af927777caa8ba52bbd66" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "c914c1f5a7ad0c8c24b80ff6ea3e7e1e" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "d5714ab9cd7c3d824ceefb2c508102f6" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "a8033ff823b8083a77823204da15b953" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "9d3cbe897cbe1d6b2d24cfb767287219" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "c15a491c7e918760c80cc1e629a78153" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "7f2f0636bb381922aed6cac349df14cf" }, { "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "49cad90b2af8ee2ea20a71ecf32fd483" }, { "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "2b34dbee9336aec297d98ce3097309ac" }, { "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "f1a9b6c228749e28a03ff458bf15e9a4" }, { "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "133d22c9e03d6f6f88e44a816d973619" }, { "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "3344e51aa33b32f113ae1c47760126f2" }, { "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "ceb789cc77b7584c373f196d8a28fe59" }, { "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 33433600, "records": [ { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 14960640 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 14970880 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 14981120 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 16414720 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 19281920 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 19292160 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 19906560 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 33013760 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 33423360 } ], "md5sum": "b6f6b397d1f65c688d94a3728c243285" }, { "dataPath": "params_shard_58.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "4327877d8048c81169611c99a835a65d" }, { "dataPath": "params_shard_59.bin", "format": "raw-shard", "nbytes": 24586240, "records": [ { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 0 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 1433600 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 4300800 }, { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 4311040 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 23971840 } ], "md5sum": "5ee28d2b651a075882d9f952fbec0281" }, { "dataPath": "params_shard_60.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "abaf27d0a31629ce87f5e1ce78aa77af" }, { "dataPath": "params_shard_61.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "e31ffc8618758ed56fe9fcede44ecc74" }, { "dataPath": "params_shard_62.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "d30aa69296082797abc982ed57aacebc" }, { "dataPath": "params_shard_63.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "cd4a9d6dee12b4dd8c28bc391744b589" }, { "dataPath": "params_shard_64.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "95c17498ffc9bcfd6db9687abe4061b1" }, { "dataPath": "params_shard_65.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "3cc8b4cf639ad2e717b592f81258c44b" }, { "dataPath": "params_shard_66.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "2b84cce311191560e3b627c43a3e4c93" }, { "dataPath": "params_shard_67.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "2b3a43828b2caad40dd9eaa1b60c8ad8" }, { "dataPath": "params_shard_68.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "b3ec9a3b20ed6d20f122b1f80afd379e" }, { "dataPath": "params_shard_69.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "fedd3c427957accb9396bfebfa226b20" }, { "dataPath": "params_shard_70.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "e2eb75dfee0847d6b816938a7936fd29" }, { "dataPath": "params_shard_71.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "5d4050072d03196ed561c24dc5a4df86" }, { "dataPath": "params_shard_72.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "3a05810b04da797ad8caf94c6229733e" }, { "dataPath": "params_shard_73.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "381888e3015fd9a27529b10644007e43" }, { "dataPath": "params_shard_74.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "86b2cf3f7589a64e6f79769679742104" }, { "dataPath": "params_shard_75.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "5fda4011e84c7107cd317a2981a0b368" }, { "dataPath": "params_shard_76.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "7f9e580a84e5520534d6d814d82559ed" }, { "dataPath": "params_shard_77.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "c01e1b8ba083a1c978b68484bdccaca4" }, { "dataPath": "params_shard_78.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "19514b2db9349ea732beec60c39b6c65" }, { "dataPath": "params_shard_79.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "3e769b6a1bf76cc84663e068e910d45d" }, { "dataPath": "params_shard_80.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.16.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.16.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "597a5b975e8402f99ce8402a45cdabcb" }, { "dataPath": "params_shard_81.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "eb6fd5eacb4cdbe2774c7455169602a7" }, { "dataPath": "params_shard_82.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "36608dd48c4803da4b19855542858d40" }, { "dataPath": "params_shard_83.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.17.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "4c0a495fcab66223299c17a92d350da1" }, { "dataPath": "params_shard_84.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "94b4a64b38973c128272094d1005bee5" }, { "dataPath": "params_shard_85.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "03a3e8d784b1c3b76702ad45f564a3a8" }, { "dataPath": "params_shard_86.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.16.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.16.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.17.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.17.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.17.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "01a0a8c06fb0b8af454a778983de7613" }, { "dataPath": "params_shard_87.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.18.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.18.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "2f5f510bf1129c014edafb298282744a" }, { "dataPath": "params_shard_88.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "bb5e76f307b422d03f2fa5cc842b0087" }, { "dataPath": "params_shard_89.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "6146907f96ed3d701802a2e95c7e7556" }, { "dataPath": "params_shard_90.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.19.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "8db3e3557f602118103f35baef26f055" }, { "dataPath": "params_shard_91.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.20.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "56594d7cc830219ec12b281d6c6e4850" }, { "dataPath": "params_shard_92.bin", "format": "raw-shard", "nbytes": 32583680, "records": [ { "name": "model.layers.18.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.18.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.19.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.20.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 31969280 } ], "md5sum": "ad3935c5fb04123cd0e899871e1933ff" }, { "dataPath": "params_shard_93.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "269e68b5aabb573c5f9efbbe61682882" }, { "dataPath": "params_shard_94.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "4103b79f56435e66f34f3d8b4520aa35" }, { "dataPath": "params_shard_95.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "389061eea7f0e4474f876147c0e2e857" }, { "dataPath": "params_shard_96.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "2edaba5246732d18c8c1f5d656c2bedb" }, { "dataPath": "params_shard_97.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.21.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "97d1926c46c5d46c50c7fce9d91b33c8" }, { "dataPath": "params_shard_98.bin", "format": "raw-shard", "nbytes": 22773760, "records": [ { "name": "model.layers.20.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.20.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17838080 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 17848320 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 19281920 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 22149120 }, { "name": "model.layers.21.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22159360 } ], "md5sum": "58f09ea5525a7d9e6741a22ca127b10b" }, { "dataPath": "params_shard_99.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "f7535c72b507c0f147d83781f2ec007e" }, { "dataPath": "params_shard_100.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "8fc5335829789746f3c539992f913d1f" }, { "dataPath": "params_shard_101.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.22.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "7d1a447f41116bfd4e9dd63af865c365" }, { "dataPath": "params_shard_102.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "68d26c327c059fe9a6ef66b70a94f21b" }, { "dataPath": "params_shard_103.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "0b05d0d573238188f1f712369f7ac412" }, { "dataPath": "params_shard_104.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.21.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.21.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.22.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.22.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.22.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "b1c99db71cc54ece33cc7d9dafa39e85" }, { "dataPath": "params_shard_105.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.23.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.23.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "1d3611edb3b0e7abcb2c99f2a6cd9152" }, { "dataPath": "params_shard_106.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.24.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "c216f6010d837adc97f04dec518a3483" }, { "dataPath": "params_shard_107.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "8bc10bc64060c63d7695891212de9e9d" }, { "dataPath": "params_shard_108.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.24.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "0312423ec94d709f46e7aa39b132d783" }, { "dataPath": "params_shard_109.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.25.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "e461b75760597cf1a3f5081483eba661" }, { "dataPath": "params_shard_110.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "e54e9982d1606599d1f2a35d4fb5875c" }, { "dataPath": "params_shard_111.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.24.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.24.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.24.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.24.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.24.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.25.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.25.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "7332ff170300a81e25c6de046abd5a64" }, { "dataPath": "params_shard_112.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.25.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.25.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "df0c7f90655e6a08e7f9f519c1da1d92" }, { "dataPath": "params_shard_113.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.26.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "acf57bb44dfa560cadaf090a750a2c3b" }, { "dataPath": "params_shard_114.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "7afeae6a0126bf8af35ac42dbb45adfd" }, { "dataPath": "params_shard_115.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.26.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "501a21536e7854018f025edec582d468" }, { "dataPath": "params_shard_116.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "88dd443d1bfd9597891112fb99d48e63" }, { "dataPath": "params_shard_117.bin", "format": "raw-shard", "nbytes": 31969280, "records": [ { "name": "model.layers.25.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.25.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.26.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.26.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.26.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.26.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.26.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 } ], "md5sum": "1106b7ead4cea94290e629be810daeb0" }, { "dataPath": "params_shard_118.bin", "format": "raw-shard", "nbytes": 23142400, "records": [ { "name": "model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.27.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2867200 }, { "name": "model.layers.27.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22528000 } ], "md5sum": "3ec4477450dadf81b496f2b313676899" }, { "dataPath": "params_shard_119.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.27.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "5fb47fb36f1dfc39069f8ae14b54bca2" }, { "dataPath": "params_shard_120.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.28.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "80c280636dad9352e7f088991bdad3f7" }, { "dataPath": "params_shard_121.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.28.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "74e8e8e310a5a7fce6f39f3904199973" }, { "dataPath": "params_shard_122.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.28.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "ab0eb04596961a9e96343561fa6fce76" }, { "dataPath": "params_shard_123.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.29.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "d7339af47bfce27bc0e1269ff3edfce2" }, { "dataPath": "params_shard_124.bin", "format": "raw-shard", "nbytes": 33433600, "records": [ { "name": "model.layers.27.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.27.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.27.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.27.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 14960640 }, { "name": "model.layers.28.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 14970880 }, { "name": "model.layers.28.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 14981120 }, { "name": "model.layers.28.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 16414720 }, { "name": "model.layers.28.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 19281920 }, { "name": "model.layers.28.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 19292160 }, { "name": "model.layers.28.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 19906560 }, { "name": "model.layers.28.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 33013760 }, { "name": "model.layers.29.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 33423360 } ], "md5sum": "16964a7fe4bacead5136b20021091d2f" }, { "dataPath": "params_shard_125.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.29.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "6bab762998024e9bf7f85d4127afa374" }, { "dataPath": "params_shard_126.bin", "format": "raw-shard", "nbytes": 24586240, "records": [ { "name": "model.layers.29.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 0 }, { "name": "model.layers.29.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 1433600 }, { "name": "model.layers.29.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 4300800 }, { "name": "model.layers.29.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 4311040 }, { "name": "model.layers.29.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 23971840 } ], "md5sum": "8f78c8d10235c5cadbc0fe89902ae780" }, { "dataPath": "params_shard_127.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.30.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "62d22b201def62f86994dad3762e00c9" }, { "dataPath": "params_shard_128.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.30.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "4b9bef2afe6ed0cdccf0c8e9c6b41c99" }, { "dataPath": "params_shard_129.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.30.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "26b58cdc5ed53c983831282502c60ca2" }, { "dataPath": "params_shard_130.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.31.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "4b4026e0597566b01908312a5f745ef5" }, { "dataPath": "params_shard_131.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.31.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "932edea92a225a133c03c83cbec02e46" }, { "dataPath": "params_shard_132.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.29.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.29.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.30.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.30.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.30.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.30.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.30.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.30.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.30.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.31.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.31.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "8b5051814d16a27d04762a424760d206" }, { "dataPath": "params_shard_133.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.31.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.31.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.31.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.31.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "a1be64de4c00a6ab9411ccfb6acafb3a" }, { "dataPath": "params_shard_134.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.32.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "ca6e8248938369a9a652242451891c83" }, { "dataPath": "params_shard_135.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.32.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "a5e088d31676a5ab76686e984cc84ecc" }, { "dataPath": "params_shard_136.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.32.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "05005a07e8ea4b3e344ef36131c07228" }, { "dataPath": "params_shard_137.bin", "format": "raw-shard", "nbytes": 45875200, "records": [ { "name": "model.layers.33.mlp.down_proj.q_weight", "shape": [ 5120, 2240 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 45875200, "byteOffset": 0 } ], "md5sum": "cf16a0521ff67d49e95ae562a2e61c03" }, { "dataPath": "params_shard_138.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.33.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "f61dd383bf7f30f733a9a2246939552b" }, { "dataPath": "params_shard_139.bin", "format": "raw-shard", "nbytes": 33413120, "records": [ { "name": "model.layers.31.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.31.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.32.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13516800 }, { "name": "model.layers.32.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 13527040 }, { "name": "model.layers.32.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 14960640 }, { "name": "model.layers.32.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 17827840 }, { "name": "model.layers.32.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 17838080 }, { "name": "model.layers.32.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 18452480 }, { "name": "model.layers.32.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 31559680 }, { "name": "model.layers.33.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 31969280 }, { "name": "model.layers.33.mlp.down_proj.q_scale", "shape": [ 5120, 140 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1433600, "byteOffset": 31979520 } ], "md5sum": "96bd08ba7460bb800ac84898ac7bc047" }, { "dataPath": "params_shard_140.bin", "format": "raw-shard", "nbytes": 23152640, "records": [ { "name": "model.layers.33.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 0 }, { "name": "model.layers.33.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2867200 }, { "name": "model.layers.33.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 2877440 }, { "name": "model.layers.33.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 22538240 } ], "md5sum": "e24a98cd7855890d0c4da0e08c310b01" }, { "dataPath": "params_shard_141.bin", "format": "raw-shard", "nbytes": 91750400, "records": [ { "name": "model.layers.34.mlp.gate_up_proj.q_weight", "shape": [ 35840, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 91750400, "byteOffset": 0 } ], "md5sum": "29c915426d2b3cbc5f378c188991e2e7" }, { "dataPath": "params_shard_142.bin", "format": "raw-shard", "nbytes": 19660800, "records": [ { "name": "model.layers.34.self_attn.qkv_proj.q_weight", "shape": [ 7680, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 19660800, "byteOffset": 0 } ], "md5sum": "81a5a8dbd65de67e7e9c26ac3d887cb2" }, { "dataPath": "params_shard_143.bin", "format": "raw-shard", "nbytes": 30515200, "records": [ { "name": "model.layers.33.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "model.layers.33.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 13107200 }, { "name": "model.layers.34.mlp.gate_up_proj.q_scale", "shape": [ 35840, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2867200, "byteOffset": 13516800 }, { "name": "model.layers.34.self_attn.qkv_proj.q_scale", "shape": [ 7680, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 614400, "byteOffset": 16384000 }, { "name": "model.layers.34.self_attn.o_proj.q_weight", "shape": [ 5120, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 16998400 }, { "name": "model.layers.34.self_attn.o_proj.q_scale", "shape": [ 5120, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 409600, "byteOffset": 30105600 } ], "md5sum": "be26bd9b7290371cdbd45681ad1b6a89" } ] }