diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,3137 @@ +{ + "metadata": { + "ParamSize": 195, + "ParamBytes": 16061046784.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 1050935296, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 128288, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1050935296, + "byteOffset": 0 + } + ], + "md5sum": "ecc06a096cc46370a911d269bbd5e89e" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "ee4a154632d2aa6ec82695ddbafbfb63" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 1050935296, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 128288, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1050935296, + "byteOffset": 0 + } + ], + "md5sum": "9b55ef82466a0cecb8803f3e967983fe" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "e299cfdb94c63c55290ad929e4eca8c1" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d8cdcb042f7a39d94713b887d5b916f1" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "6b53591e6806913e62194d0aaddf1c55" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "bff5bf7cf6f2b9e38424890fad97941a" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d3802573bac229e683f051cedc46e5c3" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d80c6661159286f5216ffdb675171fe1" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "0d4f59e188d783de0996609e0db405b4" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "294d15843280ddcd89921f3963ae0ced" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4e428b66993d59889e5e00017d310f74" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "3c0159be15cf1b57584b8e089aa59f2d" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "da7cfecb809cf4df1260c5320316fb41" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "94f165b68936c1c9d6262cb089ddf235" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "3368d8027f0576822f7df5af8169d053" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "711364dfae36037bb310a9c8cecc2c74" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "24f125b60fe99a9f293fba646d0ebd42" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "79e23f3d7ce6c26b6c88a04af6ed82fc" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "f344c3acbe3da422b407c3adcc5b2d29" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "3703bbd20d99ff0897a42d777ed560fe" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "2e5b2055569e8710e397d157c6d796cc" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d6b16914e4de0e6c54a9e4e9153f7493" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "fd59471065f91a06c80c3b5b6def68ef" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "634b5527134e6e49d0d86fa0508228c5" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "3c78b36eed963699eb885366d64dbd96" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "b9f5620c225ba265d437cb12ef5e4532" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "07acdc287c5f3a993c43c56b50f8a57d" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "318780da8705700eec67eb29285a59b5" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "9f3255ebb7004d9aae3f787f0e6927ed" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "ec3f792c415e82cc0a8420165ed2164f" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "65e9d34285f1acc4845690779d4cb420" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "b9c7303f9ebf7172eaea3dfb2a38956f" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "c7d1135b98a1406ea6bde227cc018337" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "bf67a5755b8da8c2ca8db9e93b456a7f" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "2810b4f0b5d573cd47cec1e86c7442b3" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "90095a9f1d1b61e5eca59307e21f5522" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "3b69717b3068aa2d5fbe4302e0c30ae1" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "09da3922c3d5bcb28b4c7ae1cd6642c0" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "0fbeb9e99ef80514bfa46dbc609900b4" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "5bc392f7980a15716bb170096095315e" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "3320680dc8ea2d0e6cd76d5c9aa24fb7" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "952afef482d6570924e9d23b2263571c" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "92d254453e9626504f2ab1519cb1c224" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "8b1d3e60c801e1ee7c61699b12a296e7" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "f53c07f0ed94b3b97341fd61391532d7" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "16e0b92f45eb1d0f4efafb264255e686" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d688acd969d50dfd1684a380420f60ab" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "82e1c71cde9d5c4df1264e1683b47262" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "0e273234aa85c24c21f2e6cb0ad03895" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "e99c358b4bd4ecdb0ca84b2f9b3d74a4" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "cc86dac581bdfcd639ddd6bbe9140c65" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "46600ec8024bfe1ce767d97c4a69f688" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "4f2269d428f9e510bc054de7fb569909" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "849fc496126efc64ac8ab0a561a3e948" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "1ec4b31b9d23cf97335406d97d746ddc" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "bd4d6f857899366395c687b3ed9a0bff" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "529ace6a5c4aade2f65fa479693b009d" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "4620cbfd87a9e07572274916ea981a18" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "6199f83db41c1128e0c9cd8e1e977765" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "512ef8e28305e5e5e166f8deeae9fd75" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "14c498441e1b38758f3c22531d99770e" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "01cc2e64847e4fa3c913a5ebb29dee29" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "200463564c6f43fdb9c59929fb1b002f" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "900e52fb23be63e911270d93f6ad8df0" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "416d6fccab1046af0c77a037fd85ecd9" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "5bdfb454c86fbf867f0819cbb8078d1a" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "470d14b53d09559159cfd374ac1f7c70" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "1765a6baeaec42344df4243a20816557" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "a7e6ce1d1e21fdf6ff9d53e11f76cc68" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "37bc2abebdb254018c4f790644df56b7" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9b28bb0a46827ef741166a37628913df" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c1d278e1cb4b34cb63bf13996c929b92" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "38263d822dcfd1fb390194ce0c2afbbe" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c1889c49ac58b2374598ac44f27fc402" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "5ab5fe752e4e7a3d1bbc4c2790c03c3e" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "317de9668b83b33f71a0df5ed706462b" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "f216b1918b31b70188813190663fccf5" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "aec24dfbf87691bde19219d16e1931de" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "63d47b3c72b5020e49f563954a1b8faf" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "8dee9526410ec70519d1e324a1d60667" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "42c6c5ce2cf5a9b83e64e48d87b37629" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "61cdd1b2151ed4b40634fe048f188496" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "4b83a302de3fa153ae806c774b137399" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "00e5d3319304e419f6c09ad818e12a77" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "8b363d3bc2b6e8d79a696cb96e2fc7b2" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "07d739b4c2a0f3bfa6652a10de792b4c" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "4fd9c64fca40b7a50a475347199441f4" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "c84bca616c4a9688bd63025c8b0d81d8" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "9c8d7f792747c0f340b28dee0b6a24e0" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2faa78d25f3d1059a4bd4b152d6c58b1" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "a6be7d4ab4a12716cb5d797c8d97b23e" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "158af4361381bbc3c1e02f0ca72fa557" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "a06111be4178b8872a0b3df3a7b41ee5" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d0dc953859b24377fa06a9f40d68f112" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "9fcc3dd5c9b4829f51e917ff8b738c35" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "d9f34f85144950f1b68de11465101a63" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "5dad47f20c2f55b06522e87bdea92fe0" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "269cd1065d8c768ed698df343dab67f2" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "17232eb4610869341d76db804df3f642" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "ce7fbe456bbbc56154dd77d3e8ff08c2" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "896b0cb56fd740651d3cf3f9dea2e455" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "080bd372b094619404fed9ebdf4bcdd4" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "21bf5aa0bec91cfc18959c41b77c282c" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "4a86cf6bac26ce4d14a0112e5191df74" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "5df276d49abd874fcf939e359aef821b" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d415c21d2b63f1e3518bfb75589d38fe" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "eb66a11880da238253920f4a068d26f3" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "81854121c17153449410339970187a7b" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "10b242a7f52cdbdae52f1b7621a1399e" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "4281af70e081af3642e33c7e9621fe4a" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "d42592c61a1695aa500dcbe3f79d1812" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "e7d47b710cf5f7b01bb7fcd18c206530" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "887f350e05a267120492050fdf02e15a" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "792ef99016509010d7f32daab347be17" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "24f9dce1b3068c804a89f2024488fd72" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "9b548580ec4b987755519552cc399639" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "030e5ce73825551d49ca755a8df6f446" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "bbd00e86746f5c8c6bb65f319903769a" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "16746644ed74a86f83c89dc7c6d821ce" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "b1bc622966bcbeb05dbf782f949c693e" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "77542cd9bde7d1d7cbe94d6b0ae26a82" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "2217ae7b02d6bd41d5f5a901f0db6d95" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 117440512, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.weight", + "shape": [ + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 117440512, + "byteOffset": 0 + } + ], + "md5sum": "eb083bebc505b1df45fc7c99e4b53fca" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "b99bc95577936d7447a2cd4cc4335638" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "ad4d020a19b0ad8f6cce7d662eb7858b" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.30.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "c4eaba9cd647b01b529b4a712e1c8987" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 234881024, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.weight", + "shape": [ + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 234881024, + "byteOffset": 0 + } + ], + "md5sum": "2b6472f4e43db59028df41b27e2c3345" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 50331648, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.weight", + "shape": [ + 6144, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 50331648, + "byteOffset": 0 + } + ], + "md5sum": "65442520600fc32cd2ad711c9f9b2db5" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 33554432, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.weight", + "shape": [ + 4096, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 33554432, + "byteOffset": 0 + } + ], + "md5sum": "d0a869bdd171ae88d7c4e0da62e9323f" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 532480, + "records": [ + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 0 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8192 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 16384 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24576 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32768 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 40960 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 49152 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 57344 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 65536 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 73728 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 81920 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 90112 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 98304 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 106496 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 114688 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 122880 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 131072 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 139264 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 147456 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 155648 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 163840 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 172032 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 180224 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 188416 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 196608 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 204800 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 212992 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 221184 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 229376 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 237568 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 245760 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 253952 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 262144 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 270336 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 278528 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 286720 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 294912 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 303104 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 311296 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 319488 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 327680 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 335872 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 344064 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 352256 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 360448 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 368640 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 376832 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 385024 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 393216 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 401408 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 409600 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 417792 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 425984 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 434176 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 442368 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 450560 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 458752 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 466944 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 475136 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 483328 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 491520 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 499712 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 507904 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 516096 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 524288 + } + ], + "md5sum": "29d44acb4890cf05806a9cef56b982ac" + } + ] +} \ No newline at end of file