{ "metadata": { "ParamSize": 269, "ParamBytes": 1033572352.0, "BitsPerParam": 4.501551474039708 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 155582464, "records": [ { "name": "lm_head.q_weight", "shape": [ 151936, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 155582464, "byteOffset": 0 } ], "md5sum": "5f0930c47ad64e377619c89161b663aa" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 28901376, "records": [ { "name": "lm_head.q_scale", "shape": [ 151936, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 19447808, "byteOffset": 0 }, { "name": "transformer.h.14.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19447808 }, { "name": "transformer.h.14.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19460096 }, { "name": "transformer.h.14.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25751552 }, { "name": "transformer.h.14.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26537984 }, { "name": "transformer.h.14.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28635136 }, { "name": "transformer.h.14.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28897280 } ], "md5sum": "5371b7a6a7e287b08a0c195f974cc13f" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.14.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.14.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.14.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.14.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.15.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.15.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.15.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.15.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.15.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.15.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.15.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "37d71f502de2d720ccbf13e3b371015d" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.15.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.15.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.15.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.15.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.16.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.16.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.16.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.16.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.16.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.16.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.16.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "4f3692401592b8e8a5fbde1bef304564" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.16.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.16.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.16.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.16.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.17.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.17.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.17.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.17.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.17.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.17.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.17.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "7980eef9472808e7b64af041116f1f1d" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.17.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.17.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.17.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.17.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.18.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.18.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.18.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.18.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.18.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.18.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.18.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "579a592fb66cadb1444be6548237ed13" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.18.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.18.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.18.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.18.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.19.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.19.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.19.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.19.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.19.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.19.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.19.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "3c10f26729dd19846343e9f21e55ae2c" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.19.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.19.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.19.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.19.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.20.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.20.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.20.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.20.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.20.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.20.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.20.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "5420912553ba88ad446a4b835d41f475" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.20.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.20.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.20.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.20.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.21.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.21.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.21.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.21.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.21.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.21.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.21.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "4617b1c4f7ca83e64c71f10aa2e85eec" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.21.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.21.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.21.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.21.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.22.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.22.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.22.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.22.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.22.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.22.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.22.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "6d1a5c51473dd393bcb2fd80c2aca882" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.22.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.22.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.22.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.22.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.23.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.23.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.23.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.23.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.23.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.23.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.23.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "f362dcc35078d3f4c867f5e1ea09e301" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 28483584, "records": [ { "name": "transformer.h.23.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.23.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.23.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.23.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.ln_f.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19021824 }, { "name": "transformer.h.0.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19025920 }, { "name": "transformer.h.0.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19038208 }, { "name": "transformer.h.0.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25329664 }, { "name": "transformer.h.0.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26116096 }, { "name": "transformer.h.0.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28213248 }, { "name": "transformer.h.0.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 }, { "name": "transformer.h.0.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28479488 } ], "md5sum": "024a22e34e5ec7b319d8083a87b60f7f" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.0.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.0.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.0.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.0.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.1.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.1.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.1.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.1.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.1.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.1.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.1.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "953d1ef771511ead595bc8c669fc0564" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.1.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.1.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.1.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.1.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.10.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.10.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.10.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.10.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.10.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.10.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.10.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "9e55c1655d55defc08e03f1f53f99fb5" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.10.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.10.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.10.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.10.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.11.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.11.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.11.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.11.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.11.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.11.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.11.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "e486572ccf92b5d21ce3b5679f0ba5cd" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.11.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.11.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.11.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.11.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.12.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.12.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.12.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.12.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.12.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.12.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.12.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "ae6db91f0845607cd55e78bd554bd3e2" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.12.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.12.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.12.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.12.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.13.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.13.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.13.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.13.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.13.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.13.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.13.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "31bfe020f99696bb11083543b8f0251f" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 28483584, "records": [ { "name": "transformer.h.13.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.13.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.13.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.13.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.14.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19021824 }, { "name": "transformer.h.2.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19025920 }, { "name": "transformer.h.2.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19038208 }, { "name": "transformer.h.2.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25329664 }, { "name": "transformer.h.2.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26116096 }, { "name": "transformer.h.2.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28213248 }, { "name": "transformer.h.2.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 }, { "name": "transformer.h.2.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28479488 } ], "md5sum": "e381c9c430b79af07c3ba8809f56e1d6" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.2.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.2.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.2.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.2.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.3.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.3.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.3.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.3.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.3.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.3.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.3.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "9d9905aaef1bd4351d6f986c41ba696f" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.3.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.3.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.3.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.3.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.4.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.4.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.4.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.4.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.4.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.4.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.4.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "95279dd22fa9df3b0dcb955f056c2ab3" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.4.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.4.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.4.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.4.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.5.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.5.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.5.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.5.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.5.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.5.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.5.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "7d417f7af22d7185993fb6331921932f" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.5.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.5.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.5.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.5.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.6.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.6.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.6.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.6.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.6.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.6.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.6.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "1bbd8c6682ce7ed6130d8453521b60d8" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.6.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.6.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.6.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.6.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.7.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.7.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.7.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.7.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.7.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.7.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.7.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "832a89b950981b1d2c809302e89a601d" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.7.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.7.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.7.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.7.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.8.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.8.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.8.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.8.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.8.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.8.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.8.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "e70feecba6d72e6b763400d92dcb6927" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 28479488, "records": [ { "name": "transformer.h.8.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.8.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.8.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.8.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 }, { "name": "transformer.h.9.attn.c_attn.bias", "shape": [ 6144 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 12288, "byteOffset": 19021824 }, { "name": "transformer.h.9.attn.c_attn.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 19034112 }, { "name": "transformer.h.9.attn.c_attn.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 25325568 }, { "name": "transformer.h.9.attn.c_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26112000 }, { "name": "transformer.h.9.attn.c_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28209152 }, { "name": "transformer.h.9.ln_1.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28471296 }, { "name": "transformer.h.9.ln_2.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28475392 } ], "md5sum": "8aabd29f06ca59ad232b8581feafcd69" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 155582464, "records": [ { "name": "transformer.wte.q_weight", "shape": [ 151936, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 155582464, "byteOffset": 0 } ], "md5sum": "ad0fb00af46897d301560f79b827841f" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 19447808, "records": [ { "name": "transformer.wte.q_scale", "shape": [ 151936, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 19447808, "byteOffset": 0 } ], "md5sum": "0773ce91c9bc2eda490f9005b9ba398b" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 19021824, "records": [ { "name": "transformer.h.9.mlp.c_proj.q_weight", "shape": [ 2048, 688 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5636096, "byteOffset": 0 }, { "name": "transformer.h.9.mlp.c_proj.q_scale", "shape": [ 2048, 172 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 704512, "byteOffset": 5636096 }, { "name": "transformer.h.9.mlp.gate_up_proj.q_weight", "shape": [ 11008, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 11272192, "byteOffset": 6340608 }, { "name": "transformer.h.9.mlp.gate_up_proj.q_scale", "shape": [ 11008, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1409024, "byteOffset": 17612800 } ], "md5sum": "a714d816d7b0350d4c375aaacbb392c1" } ] }