{ "metadata": { "ParamSize": 163, "ParamBytes": 695242752.0, "BitsPerParam": 4.500628909972241 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 131334144, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 128256, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 131334144, "byteOffset": 0 } ], "md5sum": "948d42e013589644f717b8f2742e703e" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "15eb3b2feeb6d68229db3ad44b4f2bd4" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 31498240, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 128256, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 16416768, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 16416768 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 16420864 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 24809472 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 25858048 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 27955200 }, { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 27959296 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31105024 } ], "md5sum": "fc88f9c86f4dcc481fc967035907410b" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2359296 }, { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2363392 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10752000 }, { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 11800576 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 28577792 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "4f796c5a5991ca5965877e22c0c35ee5" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 32116736, "records": [ { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 5898240 }, { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 5902336 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14290944 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15339520 } ], "md5sum": "1cbcc5386196039e346ebeb69898be21" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "83fd8f7f11b5e4ac336af937d7cb2e03" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 25444352, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2097152 }, { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 2101248 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5246976 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5640192 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7737344 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 7999488 }, { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 8003584 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16392192 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19537920 }, { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19542016 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22687744 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23080960 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25178112 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25440256 } ], "md5sum": "078ca4e50e6d3c20ca35afa9ded59781" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 31854592, "records": [ { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 9437184 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26214400 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28311552 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28315648 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31461376 } ], "md5sum": "75a32850b874dc2d7e1e49c7913c1a59" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2359296 }, { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2363392 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10752000 }, { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 11800576 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 28577792 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "cae69f7feb17a355e114f6ed5c41d16d" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 32116736, "records": [ { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 5898240 }, { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 5902336 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14290944 }, { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15339520 } ], "md5sum": "a948d2da6e2437e88ed87ff92555d613" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "779123f78591d201721a714b041a36dc" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 25444352, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2097152 }, { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 2101248 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5246976 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5640192 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7737344 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 7999488 }, { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 8003584 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16392192 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19537920 }, { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19542016 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22687744 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23080960 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25178112 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25440256 } ], "md5sum": "67fc262973464e45ff23270f3ca6482f" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 31854592, "records": [ { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 9437184 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26214400 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28311552 }, { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28315648 }, { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31461376 } ], "md5sum": "05ebb1fbc1a4bb5833f93c7e9db04fdb" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2359296 }, { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2363392 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10752000 }, { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 11800576 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 28577792 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "650005aa72d260141af844e88c6c6911" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 32116736, "records": [ { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 5898240 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 5902336 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14290944 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15339520 } ], "md5sum": "3ea8751f7e421b4d92f9328994ef004a" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "2c0f3cc1f421ea2b7a915cd1dc66bbd3" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 25444352, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2097152 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 2101248 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5246976 }, { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5640192 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7737344 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 7999488 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 8003584 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16392192 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19537920 }, { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19542016 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22687744 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23080960 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25178112 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25440256 } ], "md5sum": "ad9ad5ce433132937ec654f5459a2d2e" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 31854592, "records": [ { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 9437184 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26214400 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28311552 }, { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28315648 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31461376 } ], "md5sum": "c65f31110bf484ebe8c1da4e95c69670" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2359296 }, { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2363392 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10752000 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 11800576 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 28577792 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "dc2c3e2efb8e630a77a5073f88a15e18" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 32116736, "records": [ { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 5898240 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 5902336 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14290944 }, { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15339520 } ], "md5sum": "6da19bf2c8fab74a857605e061d95f36" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "2edd47f924e371584b2fe430cd82b85a" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 25444352, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2097152 }, { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 2101248 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5246976 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5640192 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7737344 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 7999488 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 8003584 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16392192 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 19537920 }, { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19542016 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22687744 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23080960 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25178112 }, { "name": "model.norm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25440256 } ], "md5sum": "309f1e4043afde5a97781e0ab5cc5683" } ] }