diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,3025 @@ +{ + "metadata": { + "ParamSize": 199, + "ParamBytes": 15231233024.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 1089994752, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 152064, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1089994752, + "byteOffset": 0 + } + ], + "md5sum": "e3ba2e16e9abcfc157668a2b7ced62ed" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "0135b2cf80ed2fb65e8549e17a1ee5ba" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "8b80efbd5bbd05274fa3302c91775881" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "d931da20410b02e97cf6d71810484bf1" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "4b679bc98ae3a2a46d8bce727dbf4406" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "62dd4f3b2d243ec10a26e002095212dd" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "6ea2fc59b5e8702c63841488219759ac" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.24.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "fbe10e280763eda563410692fd20a74f" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "306be32b1cc32e2fe9e41a03e03efa8f" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "8aee0ef3c672127ed05f1d88ec2c2487" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "7244d9719161185c3e30c294932272ec" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.25.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "5ec48089f4487d45fb4209d54107d200" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "dc77a42d1522dc38c306e9351af78a05" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "c5f25e378dbc35e5546ed174a0b8f0d7" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "d8ab08c16efe9f0dae43359eb78205bf" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.26.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "7f29306a5beedf0f52185db7f63a9f1a" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "4ab32535428babd99510ae2e35c33ddf" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "4dc4b30b6f705699bc77fa5f039ce29b" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "a591ff5675bf5fff75a90043286e102d" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.27.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "4844e51d721e59fac91f1ededf5f430e" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "21100bea6dd3e78acea0967215b73fd9" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 1089994752, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 152064, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1089994752, + "byteOffset": 0 + } + ], + "md5sum": "0b11efd86bd5a1760dfde02600a8ff53" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "b9ee0721eab7a5969496b95b8aea2da0" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "7e8c0c960fa8e03234f387caeff199ef" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.0.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "983206d0b6bf366adf2af144d3432286" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "302470ba377c815c291a60bf77fe3838" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "19fde0829560ce5a62d1297a0d347608" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "6aa02bcce79ecf8d2bce7c840ae4b9e4" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.1.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "52f19adbfcc9b9784880a740c571ad6e" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "94ccef8684bab4d4e927173cd9be9fb7" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "4f7528833d7b3b2243a174be6ff74692" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "af356424edb02a2e4bfde74eb4ea8bf5" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.2.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "5e53ff969e4c8066fafb00e0b6f4dc9e" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "5194b3fa45ed1fae1f2e0664763d5345" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "897969aad96acb3bba57ad8a669cff97" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "36abf81715bc8cca988a0949d0e6dab9" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.3.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "b978e83433115ba4cdc3220c095113e2" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "a6c79c07c2b1cf92c2bbe416acb11390" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "3f6ef247aa36ede741b7c7cf26b2b9c3" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "6b30f9134cda959ffc5c33cc9300d543" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.4.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "7549ee01256438c437ba233f5ae5e669" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "aaaa78f7e503df5eeea3a15e3d2bbfd9" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "bbb275a7bc382c169ed48c1b5a0a4319" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "b359969abce3fa0b885338d79c2ae324" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.5.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "ecf8c40cd8db6f667062519d4340d441" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "2e08d5cc21b701eee963c01a3741d5e4" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.6.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "19d54c9ea0abe873c940c94337115593" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "3766515d8310e564433a2ac22403f3aa" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "77e7a30847a98c0c0ea9c8a3c57f8e16" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "a741b754e800c92cd99672bb31ad3880" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.10.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "fd4fd66a03426a14f5c7a0b96cf6da05" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "48652dfbcfa38f236b1904cd14d05ff5" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "8d0bfc57fa3e44bb12517f868661f7ea" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "ab06cec8feee1103d314097118a7dbdd" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.11.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "904d94e32b87dc95b3abf525260dc401" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "a4d89f828181f09949803bfb96150250" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "9fe9c45cb575e9d7331efe4154d569c7" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "594a05b52b4b88802bb619759f13699a" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.12.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "c6c16c48711d0fa8a04b64dd0036b953" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "4b2627e2bd76fa6195f3d1a98923ece8" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "069ca1dcb259c5bfb8236bf5d70d2d49" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "b4db69348d8002335baad53e2b221a03" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.13.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "8ffb9fab6930b28f54ad534025372378" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "18fb7173f62f55c20d6fc531d8aeef8d" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "e84d7e84582d42e0d6015d89b0913d57" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.14.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "83e118d22a8b629359a1a61ff82a146d" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "b3dbf732f0e0634d008ead6619fb53f4" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "9c7271a153dd2a6980b86d720d5976f3" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "dc2107508fa4d0fec5dcd749b949b14e" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "b19c78b3021024fdc4d21bea6d991a9a" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "767f420a9cb30cc284f344504b901326" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.7.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "ba7e28321fa86e4711645b92b06b3a4c" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "2a2a11bfd6943a8d2737165c682fe33c" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "2d44e37c5bd47c11172035a6c90a039a" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "5cb9812fd6c0f0666fb58ce62e380c0b" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.8.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "c25ab8c36ca4f385bdb35ebd1ca32ec1" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "d866e0b29146e0e86d14e399d201d6e7" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "77da7c2a58382d015bfea93c640000f8" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "e5180af24842cde7a45dee1a05dd5941" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.9.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "49cc82b23b542b0174a2193b9236e743" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "70a31f5c11df114e97be85fd427e6451" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "6da6891d4516131225870cc180024416" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "62764e665a0ffd2053dbc85fbf898455" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "265f4926f9ed674cfa62b40ee673051c" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.15.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "ebaf01de92ce3b4fd0106cd2bc24e561" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "9e284bc3bf08d9a3f6c6778d94280752" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "d54ba53285df69dc8874fd468cad8768" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "f2688d265d6418dcb7ba6673f1d2fd0d" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 33553408, + "records": [ + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 7168 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 14336 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 21504 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 28672 + }, + { + "name": "model.layers.23.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 37888 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33068032 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33075200 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33082368 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33091584 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33098752 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33105920 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33115136 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33122304 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33129472 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33138688 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33145856 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33153024 + }, + { + "name": "model.norm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33162240 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33169408 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33176576 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33183744 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33192960 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33200128 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33207296 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33216512 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33223680 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33230848 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33240064 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33247232 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33254400 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33263616 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33270784 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33277952 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33287168 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33294336 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33301504 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33310720 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33319936 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33327104 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33334272 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33343488 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33350656 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33357824 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33367040 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33374208 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33381376 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33390592 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33397760 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33404928 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33414144 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33423360 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33430528 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33437696 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33444864 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33452032 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33461248 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33468416 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33475584 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33484800 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33491968 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33499136 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33508352 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33515520 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33522688 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33529856 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33537024 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33546240 + } + ], + "md5sum": "ee3fa5a6fcf91305955c37d8c893ca42" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "abfa085bc10b1dd380fcde5163fdd556" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "dd806e048ea27bf4b89e5d53447810b0" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "c8ded8970c72a6cb63d5603ec55e26d8" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.17.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "a9ec71e3d9081d4e95552bad3ae8b9a9" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "8adf83cddc3b20b44c6ec5466a8ad12b" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "b09c85fa4c03b49a9399417b8fe6334c" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "d4c1d369d8aa3fa9d5c7987f313c7bc2" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.18.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "f24c55b4a595caa6a4787547f2cc091d" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "249dfed1294b8747240bd27ac659708a" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "e755cfc373946fd6688e5e169beafab5" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "5542f5898f9157a27b6e7ab8e666f4db" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.19.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "eddd963ada99af5e602d1fcbdcb44a20" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "350b7400de0b8e1cbf78392c2c9a324a" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "9b5f8d245c8001a11f3946d39a12f76c" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "2d8285294ac2965b8e332cadf4b051fa" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.20.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "43ed498d1a0a93898e333707b19a9053" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "c172d8ce347373bcb73a66a6cb68c070" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 135790592, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 3584, + 18944 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 135790592, + "byteOffset": 0 + } + ], + "md5sum": "becd6e9c1126b0933eca1b277948dc07" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "4f34c1304cd4ff8990b1e4acc25c6ad6" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.21.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "776051647e0ec75b5205737be7202b52" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "5d409a0aefb4cf5c16533634ee10933c" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 271581184, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 37888, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 271581184, + "byteOffset": 0 + } + ], + "md5sum": "2778594d01c31c5f21c66dd2900633a5" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 33030144, + "records": [ + { + "name": "model.layers.22.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 0 + } + ], + "md5sum": "de288520bca28438f959d0742cd582cb" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 25690112, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 3584, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25690112, + "byteOffset": 0 + } + ], + "md5sum": "d733ca9a13311902be21e7b022aa242b" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 33173504, + "records": [ + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 7168 + }, + { + "name": "model.layers.16.self_attn.c_attn.weight", + "shape": [ + 4608, + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33030144, + "byteOffset": 16384 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33046528 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33053696 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33060864 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33070080 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33077248 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33084416 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33093632 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33100800 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33107968 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33117184 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33124352 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33131520 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33140736 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 33147904 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33155072 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 33164288 + } + ], + "md5sum": "18a8084a5082495c094a1d8214c4e903" + } + ] +} \ No newline at end of file