diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,5153 @@ +{ + "metadata": { + "ParamSize": 339, + "ParamBytes": 29540067328.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 1557135360, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 152064, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1557135360, + "byteOffset": 0 + } + ], + "md5sum": "0c1d58abe11d7c56f26ff83c02e983da" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "4f6355b80f0e5db7034e7b9497f9df83" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "7cfdb93c2ab3006d92f651b1f55cd24e" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "531068dd28c51017e890687cb215728a" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "614dddbddf20693540f6ba9f12e482a4" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.34.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "c59ed4fa453a1267c084e7b5b1e351ba" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.34.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "09196ca4c00a9ae00bb3c9277031d2b7" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "3ed03f6cedae5aff764e9d42349e68df" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "8e61096d52a34c6c70e060dd1c3d604d" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.35.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "4bdbb8fd782626491b1f5bcb41ef01ae" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.35.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "892d55e729930338f4d3022eed93d2e9" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "2a36f2e86cf2c0086dd5c0d5e9dd0eeb" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.36.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "903680202e6772f8a4f0a9af3fbb79f6" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.36.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "faf53818327900e0fff25b39e2b97fc6" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.36.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "88f1d55534e2c4d098bbe47c84b3aa82" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "854e1d796967d6cb13f81afee1eede2d" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.37.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "c56afb7bf7beddb5f1c1e3ad1c4376c1" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.37.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "1cd99c02a5ed0229f63529e3adba631f" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.37.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "aaa55b768ffedb8b96bebfb6d5171429" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.38.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "84c518dc1027cd2954f3f32cb44e98a5" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "9a49c541d21dce60eddc668a414f92e8" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.38.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "13ad74508a4197080ff3fd3226fd1760" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.38.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "ce00cd6b6a83745d3e559feef7aae33b" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.39.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "1e5b4cc51b300293b1a52f0a18ef53dd" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "45407c819744720519210df4ca60cc22" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.39.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "2e9e5cb33dfca31095dfd3bec03fce9e" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.39.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "f22f9afafafee1cba036a5abe20e2177" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.40.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "01d605bde1918aaa02a6bb1932bb3e75" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.40.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "172e9d8356bdeb5b0c360e1b6da1f78f" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.40.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "dfc1e27e140911315121a852bf6a934b" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.40.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "95a41cf0ddb042377f859be5096e71cc" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "8643dbb10efa96a6617bdcea187b0180" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.41.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "8a6e92b8edf9befbfa8900b91258973c" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.41.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "e073b58a244a7968ad333bf2558ce8ca" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.41.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "569b1354dc11cb7ccf50f29f33bf172c" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "10c0d25f8ccd4b263db1d248d8a2a22b" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.42.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "970c174a60be35841a8916248218ce1f" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.42.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "ffd12e725e94e6c7b92eb1a28fcbb424" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.42.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "5cbaca2b0939f67d45c778c40d9bfed5" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "6a0e5cdea07d24b95b576ce0a6988931" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.43.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "1b013461c4a0e24287c7b0b922515fb5" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.43.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "2a29efe7e29150235da31088f0f4f577" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.43.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "d555c6f73a3713a1a1b1c31be1264266" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "d8c8f3eae2eb03ca946627d2cf05e6b7" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.44.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "cf8754b46e1dbf91e0e9a18bb90fd586" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.44.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "6550ccf59a23379631e238e82004c0b6" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.44.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "5872fb841bd8a50e749405a3dd8e7975" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "ee6160ca1a68a1a41a076f5df9c68561" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.45.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "45f44736670cf9db26f906bcfe2eda4f" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.45.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "d1203153208f0ac3f5b25e1e6117cd9e" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.45.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "3f010deb24fc5fbd1786ac66f6530d26" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "bac37662f62450edf94e9ce2ed0f803a" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.46.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "e9bb286e88d6778dd29b35195778dfe8" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.46.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "b8b078cb1580080fd74a8be691f0ae8f" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.46.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "3c59093fea25eb61b810f863c7732907" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "4177727f3d535c87e43a582fdc78b243" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.47.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "b99de658540239876b6e375297aa1c18" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.47.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "79184afb2a28b4c7de7f399f39c37bb2" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.47.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "cc2814c41f5302b42853ba0159391510" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 1557135360, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 152064, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1557135360, + "byteOffset": 0 + } + ], + "md5sum": "d3bce2aba6a0419f849d6ca7840ea54d" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "93a1a0064719de22b653793a3c28c570" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "f6fd658a6bda0dffa39e4004204d359a" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.0.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "c5e891d7b3258d04f289bba662a5819f" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "2b1942e5cb2b052396bd31f950e64785" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "36797fbd253b45ebeccad47d7d05d588" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "61fc1d23f5a943a5134a2b8096bcf0d6" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.1.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "cf9deaf2a7932f18579d8c7c9e29a392" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "1fdb82b9de382131a7aa1ebca833199b" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "c657ec00884e495ee92812e849a86bc9" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "36bcf943f720eeff55c8fc017c4c64e7" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.10.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "ae7ee50ae202536e2c0f1a4355972a69" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "75765e2a54ed38f98074e002ab37a6f8" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "4809d8f8b134b4147b2f533733ecbda7" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "a3cbf852f621be9a7f508b86a9c82e1c" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.11.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "0fddcb8d02fbed81ae704360d8ffd9a8" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "f23e42c2ee78c62d60cbe623256fcdc4" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "8486377272eef5f5767b529d34d0efd8" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "95dbbf6217cbabd3cb4bc1002e1861db" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.12.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "180f22acaaaa59e0d350057074ea5d0e" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "4dd2682dfaf7c0836ce87af8609949a5" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "429d057b44545a0b2d44fc93f02639d6" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "43ea5551706d3a01fea9b66415f6068a" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.13.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "957ea9c2188565658efcbce5ede7d005" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "5dd2d5182b5160d9bdd581d7c9be9dbe" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "17dba517c064257459bcd568f00cf0f9" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "75be3fcb25d469c2267b67b6ed2d3c00" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.14.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "4fbe4408be775d01f2f5e019ef0b7c93" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "7d48416ac4d37628db0143c0e7f37b42" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.15.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "72ec3b4c50f909c35c1f73fa12831925" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "b5baa26c09627b38b0972c8ba818ec59" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "07c3ced521c3bdd7b3a543ffae74f42e" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "336345c83801733a0a3f5db604443f15" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.2.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "8d519a89d85128104447035801f228a0" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "25fe66c9c40ef7056808da75bbe0e58a" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "1e8790835be22df8a24096cab0e1eb08" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "0664414fa86b5c1bfecc39367041950d" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.3.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "62f70a69b00bdf4defef5951e1f7a96a" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "4d619fbdac358f232b5ee8892d210dba" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "fc1ea94ec793710761da4ad94938ca21" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "ca633873cf461877efdd591d2fa8d175" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.4.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "efafa0629352d1c1add4022202570436" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "fd9c5d47f4ed5cb6384e9243070251b4" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "3966edf2a9c6d872992cfd4b0dd94e15" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "59b5149e1279a2874582141d63c75c26" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.5.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "0fe4e5709b1e6baf3e456c77382f815e" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "23fa5c2f0f6ee962ae79e0759f17d994" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "3661e43885de6394ad3ee8030231d31b" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "bbfccec013296fd294aa6db06977f898" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.6.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "5fb4709e98b17d3dca0b6790c4cec006" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "591b8b711b4a80d36174f477344ea944" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "89e9a0e3d6affe1889f5744e0caa82d3" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "70568efc36a587ce04b20520f51b25e0" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.7.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "0d4675ce6207fa318421a439664c3313" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "95750b08256ced4e6e65bda63b856dcc" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "c6926e0cf173c045926b99b1cacb2cef" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "daef92aef3765a99d467a1c1b0e78225" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.8.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "755b2316694e6f21713b8c330b73b6a2" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "ca2b9169a53cd9e6bc8c65dd3531c1e4" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "b500d528e5b14cfcc43027b220833620" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "f8ee29c0d335464c35ac7034916e849e" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.9.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "40ff78f7bf479cf874f459353f834f42" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "64d6326154bfa54d090d5b2eea19d266" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "a501b5ad3b68369fac812706c402e323" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "9707a8d39006ff3c2ee5160bb3e992c6" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "df935f57611d30bda15bac985780ed6f" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "33a74c65c91d1fd781cb6ae74fdca7f6" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.16.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "09da739d48e9ec8d32a12c1e2f12f859" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "e6d501b06d038f96cc56a6e831ad35a4" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "76d953f26fc4cc6a55ee1a4a5e7ef299" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "2624be8633729ecb541502fbdc32d725" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.17.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "768c819a474f8ce6e041dac851986b10" + }, + { + "dataPath": "params_shard_131.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "67daece528561fc21cef0822d25a976c" + }, + { + "dataPath": "params_shard_132.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "6bc2fb030c6cfc027ad5796b3630c8e2" + }, + { + "dataPath": "params_shard_133.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "f27182350398dee076c679a922f89d98" + }, + { + "dataPath": "params_shard_134.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.18.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "ea62b2fb09717d64b0ecba1d8f7945aa" + }, + { + "dataPath": "params_shard_135.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "5dd7e4123c222143fdbe46f2e61893de" + }, + { + "dataPath": "params_shard_136.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "9ffceeb9c270bf88dc7db969a29f3588" + }, + { + "dataPath": "params_shard_137.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "7500ce8712e4797d575151cb80a80e23" + }, + { + "dataPath": "params_shard_138.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.19.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "7914bfdb914ec016621d53e562fa5513" + }, + { + "dataPath": "params_shard_139.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "6375ae3555a0e1142f457c84bb200040" + }, + { + "dataPath": "params_shard_140.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "619300082d554277cedf0e4944d5814f" + }, + { + "dataPath": "params_shard_141.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "0aa1965241dcb7bb137de8d2e9ec92a1" + }, + { + "dataPath": "params_shard_142.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.20.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "15e45fd98ff7c2f83a6ebf97b53cb919" + }, + { + "dataPath": "params_shard_143.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "db647d0bfee45f37d049a92bc662711e" + }, + { + "dataPath": "params_shard_144.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "8464e10e3bc76324f8eaa9f126997054" + }, + { + "dataPath": "params_shard_145.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "46a17a820de47aabd231aae9f81e98c5" + }, + { + "dataPath": "params_shard_146.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.21.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "4f8edec459e5d7af96ab5fe31e5c17cf" + }, + { + "dataPath": "params_shard_147.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "c8111659ef0ac93e7e58b6a35f2e35b3" + }, + { + "dataPath": "params_shard_148.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "aaf9d1bdea93aa490bbecdd4e7a3ebc3" + }, + { + "dataPath": "params_shard_149.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "caba42b4cecc3f199d3d747b940e4ee7" + }, + { + "dataPath": "params_shard_150.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.22.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "ae0b8fb31e7b99eb9d35838b8d2bf0c9" + }, + { + "dataPath": "params_shard_151.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "f04efd1c9eeedccf8198389929e6a15d" + }, + { + "dataPath": "params_shard_152.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "04f531c1c4eb0d3ec2edbb3d96994758" + }, + { + "dataPath": "params_shard_153.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "ba72755035711399057ded74ce62be1c" + }, + { + "dataPath": "params_shard_154.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.23.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "8b9ae1f52d8bbd24e7f527910457146e" + }, + { + "dataPath": "params_shard_155.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "6d476f23d22adc3b4d440f0fe2f6b75d" + }, + { + "dataPath": "params_shard_156.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "6c95dcb4add9240f9eb35aa753928bb9" + }, + { + "dataPath": "params_shard_157.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "58ca0950acfbda260987abe8919646eb" + }, + { + "dataPath": "params_shard_158.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.24.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "3a4806f67a3750a1c516d1a00b6e5b13" + }, + { + "dataPath": "params_shard_159.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "05931e29fbc75d3c36b0df7590677aee" + }, + { + "dataPath": "params_shard_160.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "eded7e470033a3f68436660f789990f3" + }, + { + "dataPath": "params_shard_161.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "07dacdbd9b8858bfb8a0a4bd57451366" + }, + { + "dataPath": "params_shard_162.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.25.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "29dccb20776f50812e4f27860a8ce665" + }, + { + "dataPath": "params_shard_163.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "eb21ba062a0670c95f964cdb365be512" + }, + { + "dataPath": "params_shard_164.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "abeab5eb345e5ea25def367b0d7f2240" + }, + { + "dataPath": "params_shard_165.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "8f169bea4a20290374d09371c75ddd90" + }, + { + "dataPath": "params_shard_166.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.26.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "24f03139366c732af52dde2331828c3a" + }, + { + "dataPath": "params_shard_167.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "664eddebe8029cbfcefdc6a0ea99f939" + }, + { + "dataPath": "params_shard_168.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "8e87b20bf9323f00b43477a3abe2582a" + }, + { + "dataPath": "params_shard_169.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "78b3f15febafa218755505305e194768" + }, + { + "dataPath": "params_shard_170.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.27.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "99e34e2357caa446f9eb0c9895c5542b" + }, + { + "dataPath": "params_shard_171.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "3ffbdcba5c6d2679994852e699fad551" + }, + { + "dataPath": "params_shard_172.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "d9bd710d0c59ef4491e8907dd8ab96c5" + }, + { + "dataPath": "params_shard_173.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "81b4203ce837d6fa1926d91c7b1f1fea" + }, + { + "dataPath": "params_shard_174.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.28.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "ed26024d733ad1b89e3dabbad4252e71" + }, + { + "dataPath": "params_shard_175.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "f0dc15e839000387e22dec5cfef4b1c6" + }, + { + "dataPath": "params_shard_176.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "a0e9440519495de4097cfe50e4a4dcc3" + }, + { + "dataPath": "params_shard_177.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "fec0edf11013ed6da18d69801ed1ac86" + }, + { + "dataPath": "params_shard_178.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.29.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "e08c7fa1b8cd68f493a79fc1080aca14" + }, + { + "dataPath": "params_shard_179.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "e3e92f43656eb39700ac845b070e9fde" + }, + { + "dataPath": "params_shard_180.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "f42b6dfefe3f8b662730e25a4b82cdf1" + }, + { + "dataPath": "params_shard_181.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "3606240c67b6f1fddbbe35f7a687cb36" + }, + { + "dataPath": "params_shard_182.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.30.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "e76877347b685ecdd28414a6ec238da1" + }, + { + "dataPath": "params_shard_183.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.30.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "78f25daf0ed04f0e2548f9f40da98f3d" + }, + { + "dataPath": "params_shard_184.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "d2a1cf60a644b5b3954a4401e36dec78" + }, + { + "dataPath": "params_shard_185.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "f13ad1d4249bca30a975eeb8d9f076ab" + }, + { + "dataPath": "params_shard_186.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.31.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "f83a8c0aa51082c743fc6849c62734f5" + }, + { + "dataPath": "params_shard_187.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "5e27b529f1bdba61e5ba0431d74f035e" + }, + { + "dataPath": "params_shard_188.bin", + "format": "raw-shard", + "nbytes": 141557760, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.weight", + "shape": [ + 5120, + 13824 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 141557760, + "byteOffset": 0 + } + ], + "md5sum": "7425074f8e7aad6d9bbfa154a3df34aa" + }, + { + "dataPath": "params_shard_189.bin", + "format": "raw-shard", + "nbytes": 283115520, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.weight", + "shape": [ + 27648, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 283115520, + "byteOffset": 0 + } + ], + "md5sum": "68442dd385cc5c3ece133b094dd9736b" + }, + { + "dataPath": "params_shard_190.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.32.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "3854a44cbe244b760c53bcd0fd64cb06" + }, + { + "dataPath": "params_shard_191.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.32.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "49f822dde835b2ec6440a6e88250aa17" + }, + { + "dataPath": "params_shard_192.bin", + "format": "raw-shard", + "nbytes": 73400320, + "records": [ + { + "name": "model.layers.33.self_attn.c_attn.weight", + "shape": [ + 7168, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 73400320, + "byteOffset": 0 + } + ], + "md5sum": "07d9d74c9adfa8966fa3a830e842602c" + }, + { + "dataPath": "params_shard_193.bin", + "format": "raw-shard", + "nbytes": 52428800, + "records": [ + { + "name": "model.layers.33.self_attn.o_proj.weight", + "shape": [ + 5120, + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 52428800, + "byteOffset": 0 + } + ], + "md5sum": "bee0869d2ed723d03bf0e1fd24462437" + }, + { + "dataPath": "params_shard_194.bin", + "format": "raw-shard", + "nbytes": 1681408, + "records": [ + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 0 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 10240 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 20480 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 30720 + }, + { + "name": "model.layers.34.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 40960 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 55296 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 65536 + }, + { + "name": "model.layers.35.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 75776 + }, + { + "name": "model.layers.36.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 90112 + }, + { + "name": "model.layers.36.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 100352 + }, + { + "name": "model.layers.36.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 110592 + }, + { + "name": "model.layers.37.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 124928 + }, + { + "name": "model.layers.37.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 135168 + }, + { + "name": "model.layers.37.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 145408 + }, + { + "name": "model.layers.38.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 159744 + }, + { + "name": "model.layers.38.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 169984 + }, + { + "name": "model.layers.38.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 180224 + }, + { + "name": "model.layers.39.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 194560 + }, + { + "name": "model.layers.39.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 204800 + }, + { + "name": "model.layers.39.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 215040 + }, + { + "name": "model.layers.40.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 229376 + }, + { + "name": "model.layers.40.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 239616 + }, + { + "name": "model.layers.40.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 249856 + }, + { + "name": "model.layers.41.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 264192 + }, + { + "name": "model.layers.41.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 274432 + }, + { + "name": "model.layers.41.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 284672 + }, + { + "name": "model.layers.42.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 299008 + }, + { + "name": "model.layers.42.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 309248 + }, + { + "name": "model.layers.42.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 319488 + }, + { + "name": "model.layers.43.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 333824 + }, + { + "name": "model.layers.43.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 344064 + }, + { + "name": "model.layers.43.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 354304 + }, + { + "name": "model.layers.44.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 368640 + }, + { + "name": "model.layers.44.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 378880 + }, + { + "name": "model.layers.44.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 389120 + }, + { + "name": "model.layers.45.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 403456 + }, + { + "name": "model.layers.45.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 413696 + }, + { + "name": "model.layers.45.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 423936 + }, + { + "name": "model.layers.46.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 438272 + }, + { + "name": "model.layers.46.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 448512 + }, + { + "name": "model.layers.46.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 458752 + }, + { + "name": "model.layers.47.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 473088 + }, + { + "name": "model.layers.47.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 483328 + }, + { + "name": "model.layers.47.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 493568 + }, + { + "name": "model.norm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 507904 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 518144 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 528384 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 538624 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 552960 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 563200 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 573440 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 587776 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 598016 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 608256 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 622592 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 632832 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 643072 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 657408 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 667648 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 677888 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 692224 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 702464 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 712704 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 727040 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 737280 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 747520 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 761856 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 776192 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 786432 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 796672 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 811008 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 821248 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 831488 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 845824 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 856064 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 866304 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 880640 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 890880 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 901120 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 915456 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 925696 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 935936 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 950272 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 960512 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 970752 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 985088 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 995328 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1005568 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1019904 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1030144 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1040384 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1054720 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1064960 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1075200 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1085440 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1095680 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1110016 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1120256 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1130496 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1144832 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1155072 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1165312 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1179648 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1189888 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1200128 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1214464 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1224704 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1234944 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1249280 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1259520 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1269760 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1284096 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1294336 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1304576 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1318912 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1329152 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1339392 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1353728 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1363968 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1374208 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1388544 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1398784 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1409024 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1423360 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1433600 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1443840 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1458176 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1468416 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1478656 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1492992 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1503232 + }, + { + "name": "model.layers.28.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1513472 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1527808 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1538048 + }, + { + "name": "model.layers.29.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1548288 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1562624 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1572864 + }, + { + "name": "model.layers.30.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1583104 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1597440 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1607680 + }, + { + "name": "model.layers.31.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1617920 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1632256 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 5120 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 10240, + "byteOffset": 1642496 + }, + { + "name": "model.layers.32.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1652736 + }, + { + "name": "model.layers.33.self_attn.c_attn.bias", + "shape": [ + 7168 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 14336, + "byteOffset": 1667072 + } + ], + "md5sum": "ade1dc870a8088f83b19f0f34004153f" + } + ] +} \ No newline at end of file