diff --git "a/tensor-cache.json" "b/tensor-cache.json" new file mode 100644--- /dev/null +++ "b/tensor-cache.json" @@ -0,0 +1,3553 @@ +{ + "metadata": { + "ParamSize": 227, + "ParamBytes": 93405585408.0, + "BitsPerParam": 13.195953996928585 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 262144000, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 32000, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144000, + "byteOffset": 0 + } + ], + "md5sum": "f590ab20b449d98958e18e40ff71ab28" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.30.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "e1abfe8addbfbca6a915f182cf433ed9" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.30.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "7095e225c03e8e9f3fdafa113cedef0f" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.31.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "0e6ec53409003a1a7a9cdee72c7c07b2" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.31.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "82801bed9caf3fb2f0f9753326c5ebc8" + }, + { + "dataPath": "params_shard_5.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": "8ccc87aa7d3cac9e60fc1cf986356cf5" + }, + { + "dataPath": "params_shard_6.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": "e64b289578fc8de8c2898df096b8ba88" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 262144000, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 32000, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144000, + "byteOffset": 0 + } + ], + "md5sum": "36cdb38eca678471e4831205151d8905" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.0.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "5bc06c92ad0fc27f29f6248e02c95a84" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.0.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "4ace676407a45a6173729b03725c999c" + }, + { + "dataPath": "params_shard_10.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": "2254ba8b3baab8bad5113ff5e9ea1d21" + }, + { + "dataPath": "params_shard_11.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": "03125726c1c5eb1d8fdb51c7bdfa5797" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.1.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "17bf9faff2f02710ff6dc2abc62155c7" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.1.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "4fac035d87b0d8f4d6481d4f0b817ff1" + }, + { + "dataPath": "params_shard_14.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": "93363e217d3713351e1a19184aa05129" + }, + { + "dataPath": "params_shard_15.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": "fb36fc504e23be7055a01fbf082e50e7" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.2.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "c4c7cf8aa7b7d629ec623cb24001402b" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.2.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "67b58ee214fb2a19c26049bc169afda1" + }, + { + "dataPath": "params_shard_18.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": "cd22cbfb13e25554f6482df035f1501a" + }, + { + "dataPath": "params_shard_19.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": "072618901a916c40dcb04d9d323a7f34" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.3.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "4c1a4ef2dfc2f004d9807852d0faa1c4" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.3.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "f78d6b247af4825dd8f5588c35d05510" + }, + { + "dataPath": "params_shard_22.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": "facb0972313dfd263d59bb552993d02e" + }, + { + "dataPath": "params_shard_23.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": "c53be52b2e620c319c5e2777e52c1d00" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.10.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "e13ad92b5ef4d955e437b570a949a2e3" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.10.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "ae5f903bcc7637b600b498fd48007de4" + }, + { + "dataPath": "params_shard_26.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": "6f9e51bddaa87ea24a1cc5a87fb28d33" + }, + { + "dataPath": "params_shard_27.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": "13d769d2b3d8945fa64e1a8392deb850" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.8.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8a2c83ef3d137a0b1a5b1a33310e9e14" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.8.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "db71f1d86e0533f4475e53e39458219e" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.9.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "98d8cc98b5a977059ff0f6b986ae4efd" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.9.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "856e94bd4b12d0e0eb4cfc6ae40c056f" + }, + { + "dataPath": "params_shard_32.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": "1099600a39151763f12d65476f1cac8f" + }, + { + "dataPath": "params_shard_33.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": "e086dfb080b7bd69c896610ddc47b2f0" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.11.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "2d76fcfbd9eb67958964951e734341b6" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.11.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "b8d82d8b715097979d95d7e743375f45" + }, + { + "dataPath": "params_shard_36.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": "05da531e857a7d18d8df10a4f12c7e1e" + }, + { + "dataPath": "params_shard_37.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": "21a09696671f0bf4c2208755f0869e6b" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.12.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "c6666d291db963137266556fdb7e7de4" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.12.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "78a6e06a3567eea1c6747d5943cfcbed" + }, + { + "dataPath": "params_shard_40.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": "baa7f8b164fced84c7428bb0ba5fbd4c" + }, + { + "dataPath": "params_shard_41.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": "d6a77d6b7d535e555988e73bda7ac4f5" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.13.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "09c1a95155fcd7c32b8e03a76fd1c085" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.13.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "5ee7f0464301c82b68d90a7b467c404f" + }, + { + "dataPath": "params_shard_44.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": "03dc286dd07f638ddc39714d927f8ea7" + }, + { + "dataPath": "params_shard_45.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": "63335693a43f621ee219698b7993980b" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.14.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "206ea5d63281f57155bcaa200d06c590" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.14.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "e32d2758d46590a5725b8b76e9e9512b" + }, + { + "dataPath": "params_shard_48.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": "2305a4686af9e8b6467db92f80b32e92" + }, + { + "dataPath": "params_shard_49.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": "bbbeab165c8b563552046da72b52c719" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.15.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "ea9dde7c434ee317a64ec69b06059083" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.15.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "db02e4f309e2f568d264ccda452df894" + }, + { + "dataPath": "params_shard_52.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": "ba1425d3518bfc0245a308531160ffa5" + }, + { + "dataPath": "params_shard_53.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": "f746f6e1f11f735d7c1bd750c2683a3b" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.16.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "5a19e7b082baba89e14c9f559381d09c" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.16.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8118cbc0babfb2d7f0e16884794bcf0c" + }, + { + "dataPath": "params_shard_56.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": "20c5ec59f0d4f589dd2c4686f2ed5fb6" + }, + { + "dataPath": "params_shard_57.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": "3c1a337f36421fc643d445f553c7c719" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.17.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "7bc07bce64a862ca7f6be49b76f682ee" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.17.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "99d5b72232173011de06735aa40695c1" + }, + { + "dataPath": "params_shard_60.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": "7fe9c9fc163326ce07c6eef5934f6902" + }, + { + "dataPath": "params_shard_61.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": "dda03f8a7bee2e58a6f5d37fb1813423" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.18.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "b772db2e4481c225647b61ac2f59de1e" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.18.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8f35ec3490b1bfd66b711742c6fb9f53" + }, + { + "dataPath": "params_shard_64.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": "4318fad29101c287b894265a37428329" + }, + { + "dataPath": "params_shard_65.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": "dfeae40956af487fd0b720e8471908f0" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.19.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "63ceeab5c9aa0fd1c0b31697ebe822f6" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.19.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "41589d8cc7de0ce5e3c167c08458ef69" + }, + { + "dataPath": "params_shard_68.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": "4119e7c6aeea54bd5b085e7ec77c3e9e" + }, + { + "dataPath": "params_shard_69.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": "f74281cfbd980b4587d1287cc33a1863" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.20.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "fe8ed0ea3369e19599fb72ffda6c4dff" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.20.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3766450f51799662fff17d51087fb7d1" + }, + { + "dataPath": "params_shard_72.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": "9dc6e10d405c489e3f686f6eb3108cd5" + }, + { + "dataPath": "params_shard_73.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": "037c9403ef3c1f7b4ddff943d71c0aa5" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.21.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "144a773efee1d7e6abd1fae9bbf8a143" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.21.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "b5ef3f509d0f3701012c5c2b39b6dabc" + }, + { + "dataPath": "params_shard_76.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": "a871cace601253e442ff7308b704f75e" + }, + { + "dataPath": "params_shard_77.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": "7ba7dc3e276b48ebeb4d1f5e16a594c0" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.22.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "fbf12d25bdab79d674bd2d00435dbd7a" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.22.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "304245f2ef2762250fc495fa071e684b" + }, + { + "dataPath": "params_shard_80.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": "b94041f1c6829a3b983b1c3a485c0fa4" + }, + { + "dataPath": "params_shard_81.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": "3a72bef383a8a79249cf64c266778d97" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.23.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "fb24412c6f24904f205d33d84a6e96b3" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.23.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "f3321725a06c58c92527cf8324f17c51" + }, + { + "dataPath": "params_shard_84.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": "a6660a6421902c3d18f6970c753f4e58" + }, + { + "dataPath": "params_shard_85.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": "31d4599c7f0f3f40a2c0f4b8aabb2936" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.24.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "aae895669db408756c6bf0d2627617f4" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.24.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "4ffdf5e2751f50da9a15ae7a596e0f45" + }, + { + "dataPath": "params_shard_88.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": "f46592bb46a00d482f00a41645b0cd06" + }, + { + "dataPath": "params_shard_89.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": "f8cd40d390334a0f599600a1e5bc9623" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.25.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "f3611b5de88a2d09ad7fad9c6bb9e86a" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.25.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "acf941847bfcc32e3e0aada0eb8551da" + }, + { + "dataPath": "params_shard_92.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": "427deb0bb74ae38311584da94a09a970" + }, + { + "dataPath": "params_shard_93.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": "370d3612cb81bc6f00e2356019989a3c" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.26.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "be548dd562cb0d4642e30d756807cab9" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.26.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "eddedd1be1c41fa50427bbee39cafe8d" + }, + { + "dataPath": "params_shard_96.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": "c3219d0c0b769fdf475d1190cf9655b0" + }, + { + "dataPath": "params_shard_97.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": "c3b1c633813726a6b9a2d2168d74067d" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.27.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "30c8787c409efcdbf0b8db5a1a033b18" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.27.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "d099a15f75f47744ff445f94a1906651" + }, + { + "dataPath": "params_shard_100.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": "faa89f12ce3432c82ca9bd03ea36821f" + }, + { + "dataPath": "params_shard_101.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": "6e29b05dbda36816545e6882fffa7ac3" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.28.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "2ae6c45bd7daf062e8775ab8cf8a1b68" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.28.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "26cdde5b1f6288f9c559eecdd76ff2b3" + }, + { + "dataPath": "params_shard_104.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": "c2cdb4dc35e15056e0a2fb47f59ec01f" + }, + { + "dataPath": "params_shard_105.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": "d3c9834141f471acb8ec9a233a67b3be" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.29.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "7d57fc0e1679fba838b990128d94be82" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.29.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "2f4a442f7fba6ca8744fc17979a7cc3c" + }, + { + "dataPath": "params_shard_108.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": "dcf68ee47c29590d68182ac9c3824162" + }, + { + "dataPath": "params_shard_109.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": "3e3c4fda2288efdf54c38996e96e9bcf" + }, + { + "dataPath": "params_shard_110.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": "05d397eccefbe41378806253e621a52a" + }, + { + "dataPath": "params_shard_111.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": "de413f5d30b35ad385b27957726871fa" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.4.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "30e216abc62bb6d2d35c8d42eef69c45" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.4.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "233abbd78796de7746b526ae24f3e602" + }, + { + "dataPath": "params_shard_114.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": "743fce880682feb6a8c028229c5974ff" + }, + { + "dataPath": "params_shard_115.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": "023c888b2861c3db3ac02783485ec955" + }, + { + "dataPath": "params_shard_116.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": "d3c64b6a1881e3c938854c421fb10d7c" + }, + { + "dataPath": "params_shard_117.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": "04b2b11bc52aab588f543b5f424e8b9a" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.5.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "6ffb3b0544ea667598ce3490ccf6c1b7" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.5.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "e4fef917e933c068bde977b5b0393ac7" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.6.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "e2e80544566fe6fbef12401fcd5bd068" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.6.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "dcd87d17393f44dcdc4a4e16eb7450cf" + }, + { + "dataPath": "params_shard_122.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": "e7d350814caf766629c27aacc1bb7ddc" + }, + { + "dataPath": "params_shard_123.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": "60eeca11650e22702c5bcaa7de48f65d" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 1879048192, + "records": [ + { + "name": "model.layers.7.moe.e1_e3.weight", + "shape": [ + 8, + 28672, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1879048192, + "byteOffset": 0 + } + ], + "md5sum": "4a279add2459687bfde379d7fe9bd46c" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.7.moe.e2.weight", + "shape": [ + 8, + 4096, + 14336 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "b30c8b8ad044a4f27945ad55160e64e0" + }, + { + "dataPath": "params_shard_126.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": "086c01d5e3f79cdcab5ee2f5a48bd9f1" + }, + { + "dataPath": "params_shard_127.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": "33a264ff9b5449df2992d245c79f0b2e" + }, + { + "dataPath": "params_shard_128.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": "fb8271758fb10263da8f9efcec4e3623" + }, + { + "dataPath": "params_shard_129.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": "58391b113b231dd3fbc1d5812ea1da75" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 2629632, + "records": [ + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 0 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8192 + }, + { + "name": "model.layers.31.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 16384 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 81920 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 90112 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 98304 + }, + { + "name": "model.layers.0.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 106496 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 172032 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 180224 + }, + { + "name": "model.layers.1.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 188416 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 253952 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 262144 + }, + { + "name": "model.layers.2.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 270336 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 335872 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 344064 + }, + { + "name": "model.layers.3.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 352256 + }, + { + "name": "model.layers.10.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 417792 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 483328 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 491520 + }, + { + "name": "model.layers.9.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 499712 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 565248 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 573440 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 581632 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 589824 + }, + { + "name": "model.layers.11.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 598016 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 663552 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 671744 + }, + { + "name": "model.layers.12.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 679936 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 745472 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 753664 + }, + { + "name": "model.layers.13.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 761856 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 827392 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 835584 + }, + { + "name": "model.layers.14.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 843776 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 909312 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 917504 + }, + { + "name": "model.layers.15.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 925696 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 991232 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 999424 + }, + { + "name": "model.layers.16.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1007616 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1073152 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1081344 + }, + { + "name": "model.layers.17.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1089536 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1155072 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1163264 + }, + { + "name": "model.layers.18.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1171456 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1236992 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1245184 + }, + { + "name": "model.layers.19.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1253376 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1318912 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1327104 + }, + { + "name": "model.layers.20.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1335296 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1400832 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1409024 + }, + { + "name": "model.layers.21.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1417216 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1482752 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1490944 + }, + { + "name": "model.layers.22.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1499136 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1564672 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1572864 + }, + { + "name": "model.layers.23.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1581056 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1646592 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1654784 + }, + { + "name": "model.layers.24.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1662976 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1728512 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1736704 + }, + { + "name": "model.layers.25.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1744896 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1810432 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1818624 + }, + { + "name": "model.layers.26.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1826816 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1892352 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1900544 + }, + { + "name": "model.layers.27.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1908736 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1974272 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1982464 + }, + { + "name": "model.layers.28.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 1990656 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2056192 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2064384 + }, + { + "name": "model.layers.29.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2072576 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2138112 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2146304 + }, + { + "name": "model.layers.30.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2154496 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2220032 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2228224 + }, + { + "name": "model.layers.4.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2236416 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2301952 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2310144 + }, + { + "name": "model.layers.5.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2318336 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2383872 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2392064 + }, + { + "name": "model.layers.6.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2400256 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2465792 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2473984 + }, + { + "name": "model.layers.7.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2482176 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2547712 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2555904 + }, + { + "name": "model.layers.8.moe.gate.weight", + "shape": [ + 8, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 65536, + "byteOffset": 2564096 + } + ], + "md5sum": "6e31beaa0ab638b43792ce1160629302" + } + ] +} \ No newline at end of file