{ "metadata": { "ParamSize": 51, "ParamBytes": 5591146496.0, "BitsPerParam": 16.0 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 1050673152, "records": [ { "name": "lm_head.weight", "shape": [ 128256, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1050673152, "byteOffset": 0 } ], "md5sum": "5844057964e6313b65a696539d583122" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 1050673152, "records": [ { "name": "model.embed_tokens.weight", "shape": [ 128256, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1050673152, "byteOffset": 0 } ], "md5sum": "10eaad0955c117845f683a4b62f09f56" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.0.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "70d13c0b62a5fc94c592e74ac5b35b80" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "dc871598d58ee3d3ab528285ddcedc3a" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.0.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "4a930acc713b0e73378198641a542695" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.0.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "914eb21e011942a506c682fb6f7d0201" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.1.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "309a07e9dc3c0e7d974271eed605f5c6" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "6aa7481b566ec48e22497ca0de7f8a93" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.1.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "bccc015f81dc7573288b330367941130" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.1.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "012af591ba86c8cdb9521e3156fbb81a" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.2.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "b3466c9b747a23c4f6709127031aea8e" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "aa0a0d5f15cfe2f9627abb16e0b0868f" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.2.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "0a48a112c40dcee2b18566f4f1371b53" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.2.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "b48376c78ea559385cedf13a7277ac35" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.3.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "2e625832e9c5547ea50cae4d36a11f2a" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "0c1f764913252c89e1c2358295700b75" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.3.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "8c464bd1ae3888da1bdcaf769cb09aba" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.3.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "f3c908cc31fe5d450e4126d2f496dcb5" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.4.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "8c1e0da9f884bdc719607fb6db4c13ed" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "e54dfe610101c525f3ea62994dbf548b" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.4.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "3ad3a17b6542c903fae96b22f58aa107" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.4.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "620cb4d0ad7824ba310ee2a40b6e9814" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.5.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "21609d866dc77f08041b363bcb86a82c" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "31eec7b07832ece340adbff4f4a7910b" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.5.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "25d0ba2ac5268bf8453c91572d8ee606" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.5.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "fa1f744d55521a9399a75ee1b691b755" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.6.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "da317bea9e7501d74284e81069098ac3" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "7a5881d0f3676263d23af8bc3dfb7391" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.6.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "1c7dfb00a746ced9777a9c575d3af2b3" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.6.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "bfada3dfaab0805b91041c03dc833f3f" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 117440512, "records": [ { "name": "model.layers.7.mlp.down_proj.weight", "shape": [ 4096, 14336 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 117440512, "byteOffset": 0 } ], "md5sum": "9b904ae6e76ed9551ed3c03a4f660a53" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 234881024, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.weight", "shape": [ 28672, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 234881024, "byteOffset": 0 } ], "md5sum": "0faece075f3aff4c43f90c0e80a61c44" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 50331648, "records": [ { "name": "model.layers.7.self_attn.qkv_proj.weight", "shape": [ 6144, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 50331648, "byteOffset": 0 } ], "md5sum": "c08ed6c1a6b260cb0e34c804f1902466" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 33554432, "records": [ { "name": "model.layers.7.self_attn.o_proj.weight", "shape": [ 4096, 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 33554432, "byteOffset": 0 } ], "md5sum": "7db25bfc49338ff6198c71217ffe9b97" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 139264, "records": [ { "name": "model.layers.0.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 0 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 8192 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 16384 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 24576 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 32768 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 40960 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 49152 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 57344 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 65536 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 73728 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 81920 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 90112 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 98304 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 106496 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 114688 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 122880 }, { "name": "model.norm.weight", "shape": [ 4096 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8192, "byteOffset": 131072 } ], "md5sum": "1ae64e59bdc4330766f1b4e6103ae129" } ] }