{ "metadata": { "ParamSize": 283, "ParamBytes": 1807423488.0, "BitsPerParam": 4.500626782697164 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 197001216, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 128256, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 197001216, "byteOffset": 0 } ], "md5sum": "7e77bcae2f042139e5607f67c99af5a1" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 24631296, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 128256, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 24625152, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 24625152 } ], "md5sum": "5ac94bd3abfff44fc23c41e36d66afaa" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "679b0a8e947d3c0eb5dabdcff8201667" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "98a4ed33191edf05b17e80e97a817585" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "4e1fa3443828c8fdeb39f564393e8bb7" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "737aa28ae0a726cbcf1eabd0985617a2" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "e3cb7516a70bb06f21e623f5024d116b" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "6efd28810753f60f93b13c31b2b8ef32" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "85c05cdc51329e47792389f287b973a3" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "62cd3323de9f04995543adcaeeacace6" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "acbad813c211c967a16de245554895e1" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "b7b3e836260920368f4a0fe22c60f97d" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "7d6fa387941d7507361cb13fec735d02" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "3c2787d3419dc5821579838c742f1239" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "c5f7101a0ef59c2215d3eaf3ec14954d" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "6929bb1f3888425307072b4f794a048e" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "b1be3539fdd9402d16713ee212d2cb29" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "c895a366991f733ee6b77f46c0b64eca" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "52383e66087a34fa6d86eb5228cb05b4" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.16.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.16.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.16.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.16.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "cfe22977e7bd8ccb8010c361c5eaf04f" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "2755836074133504271075259749a3d1" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.17.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.17.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.17.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.17.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "fa0a85839d520af38e53afe99e383b4a" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "2472e6755a42771bc1f6444362b85758" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.18.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.18.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.18.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.18.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "cc0caad6f4d3c6eeb50484e103c283b2" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "aad70fe76c46c332e21a07eeea5306fb" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.19.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.19.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "d92804441449a185f13791760c35c13a" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "599994672f4f905192a44c700fea8be7" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "0f24846c9279a0cdd4bf95b3b0c31f62" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 31463424, "records": [ { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 } ], "md5sum": "68025705e542b9200ffca14f6929622f" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "a9a21e802828f04144ae6d508165124a" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 31463424, "records": [ { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.20.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3145728 }, { "name": "model.layers.20.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11010048 }, { "name": "model.layers.20.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11993088 }, { "name": "model.layers.20.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16711680 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17307648 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29890560 } ], "md5sum": "53f92a029b6ba9f61a78d8be62ce1a2a" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "db44145d919836b013295bd12d9e78fe" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "4e90470d48bb32654445edba3362486e" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "be5210b6cb402f871085c11f301d73fb" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "54ebc0eddee00796a155bda9c4a235ad" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "82e95fb33264286664f41da47e7c652a" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "07a9e346df4c4dc6995c41cf304c8b61" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "be18d5888464c43b69b6fb86b04e4315" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "75fc1a7827e27b01ccfdeedf089b006e" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "9bb58920579b73ebd5f726bfc9863323" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "9e4f0c5818864d5de2d0b8dba9cde577" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "d9966be7afc7d6991c418f1ff7adbd42" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 } ], "md5sum": "9c1b27114c09c0d9ca2b36c8ec445367" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 31481856, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 3145728 }, { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 3151872 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 11016192 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 11999232 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 16717824 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17307648 }, { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 17313792 }, { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 29896704 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31469568 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31475712 } ], "md5sum": "2698045eda32a3475ae47de3dcf7185b" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "70b9be8dc77ffcc345c51300e1360675" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.21.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.21.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.21.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.21.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "17ccaf6cefb9cb14489198a46fd41a7e" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "9d17d4984f59f4334598aa7dc418f32f" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.22.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.22.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.22.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.22.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "319951e15b1b02a3ef2915d4fe441fa7" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "bcd3a26761b1abd783cb8075d7688cf4" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.23.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.23.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.24.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "e5d8d88f6b30f3fee5dddb77f5223485" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "7ed835e11016f4688585714ebc56e732" }, { "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.24.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.24.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.24.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.24.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.24.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.24.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.25.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "9596f0f7b37893d30e217a807750a1da" }, { "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "f00ca0cd4ae7401600165a18e68bca84" }, { "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.25.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.25.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.25.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.25.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.25.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.25.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.26.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "d65e77fd20dd9a111c2256f3b6109a02" }, { "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "57f5cb5eb3c157ab2cea9edd2297ce9a" }, { "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.26.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.26.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.26.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.26.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.26.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.26.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.layers.27.input_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "9b766eb43ec155c426d7569035610b63" }, { "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 25165824, "records": [ { "name": "model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 16384, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 25165824, "byteOffset": 0 } ], "md5sum": "c60080554e617a9d162aaa7029f09a6d" }, { "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 31469568, "records": [ { "name": "model.layers.27.mlp.down_proj.q_weight", "shape": [ 3072, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 0 }, { "name": "model.layers.27.mlp.down_proj.q_scale", "shape": [ 3072, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12582912 }, { "name": "model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 16384, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14155776 }, { "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 17301504 }, { "name": "model.layers.27.self_attn.qkv_proj.q_weight", "shape": [ 5120, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7864320, "byteOffset": 17307648 }, { "name": "model.layers.27.self_attn.qkv_proj.q_scale", "shape": [ 5120, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 983040, "byteOffset": 25171968 }, { "name": "model.layers.27.self_attn.o_proj.q_weight", "shape": [ 3072, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4718592, "byteOffset": 26155008 }, { "name": "model.layers.27.self_attn.o_proj.q_scale", "shape": [ 3072, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30873600 }, { "name": "model.norm.weight", "shape": [ 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6144, "byteOffset": 31463424 } ], "md5sum": "7dc568059e0add6ef386fb4d448e65d9" } ] }