diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,7745 @@ +{ + "metadata": { + "ParamSize": 483, + "ParamBytes": 141107412992.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 2101346304, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 128256, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2101346304, + "byteOffset": 0 + } + ], + "md5sum": "919c242370457d2349fa654fa2dd2131" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "f8b35b85370026eab7cefd0b4587c6f7" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "96c3f7e502c97f83031c426b4bfb6440" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "5ad43004d75fc3289c5cb3b9065715a8" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "82948630e6d4d7c9dc096d536c30ca81" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "1b0234c888295a6c8a94d4824f6445d3" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "708ad8c5e809ff5c72385e8aed930e08" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3126ebdb02838f9c659ad35505a51f7c" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "ce7f940dd4e2e1cd8d53d93d2c3a30ac" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "d81ed57d81433219f2de1b602e6d40e4" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "25658cfc429646cd909a64266ed3764c" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "4aa3539ddb85029976e3bccc30148a86" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "6c940514655d1424a3522a6fcb83cf3b" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "30bee48d5a4d507f1d698da166f185d8" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "31fffea396781e69f35f3e64e2995a9b" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3bcc3efcf12476b07bb2b04a712fdf22" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "f0f6ceef6084f444e9876e7ce8f88786" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "21c7d605c3775658ec584e26c5f2da07" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "b04e31de2011399457c349a09fee62ee" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "1380d996a35c501b6d4a2081250a6a47" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "1dc2eefeeae6b02fd62e10a96c3c6ef1" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "d5b33a5886dfaaa423ac50f24821b2b0" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "d2d0f492ef85ef2a8bf686c394b11548" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3aa3d7b89f8480e35d957e92c40e2a35" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "f76cddcc63f4bcd134b3a4123950e21d" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "31d0ba610470c1c485114bb205db042c" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e432ac01b027359e7231a25ee8367e70" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "865f655de1414e6dba478509811ff3d5" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "44e8be467d50a37f71cb95ccbe50ca80" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "41d25754cbcb23f20815963e11622bf3" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e4f9afcf1aaff64579f685b318ccd8b1" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "ce8b89b966e3c9876e527e392c53f43b" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "51bf12e1a7810b3cb94412bd2c5967e4" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "e4e8ba366586eee0561f352dfdd005ca" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "16c80b0a246ef04e6ef733f7efbbff12" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "678db7888782a08a0f0100d5c472da1f" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "6b3b27ac67be1e8cd9fcf7b167cce7cb" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "f8756a04dd4bc58456ad4815998516fd" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "7a06af884131d3e11a0581e80d129f7c" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "00c14937da3686671c9406efab0222c9" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "3cd6ca1e264b00495fab7b38b6084153" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "ecf9c7b8207bde2c89402ed55ee5dd2b" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "c01cadad9048f4c25ea6b3375f7a2933" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8612e35378e4830b5be4e7a71969c150" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "474eeb36fc7e48660142ae4fa0a5f876" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7c0123b391bb00114f461428fb709b20" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "4c35ed91d518d92576c156bb06dd8dd6" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "c8ad86e0d95e0ab8277b3d26a25be38b" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "7ae81584003fb5c6be8adde0313def3e" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "87e8f3f445b9b4278a4d40de0d7cb49e" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ff39ae8ec2e34cc3aa14f3845b85121a" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "515bc96756ccfe7aea5845a52fc7d166" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c5cac4c9252f0ff5db9bc99a417d518b" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "8195a4271ca4fbeb56a77417435beb4c" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "6246d204c46abc0f02e2ec37e3755237" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3c65a81a51e66da1712e552215076d21" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "177bcbfcb9ac6f05fef181468aff3d51" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "037fe63098b5312bf12f0305e69e3f0b" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "c87fe034045db936481d42a5661e83f4" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8aaa6a47e955c534404d24ffdc29a24a" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "a193005f5a1f2a52e4ef89c9ec9544b9" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "6c0ef1d446cfb5b35e60fac932377b3c" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "a74db2f784f2d344a700b3ae8f655a02" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8f3476f37edb84f3936fab9d8cbac596" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c4c2a10ffc3e326271411b8281a1dfbc" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "563f09655991a5264617eff5eb86e9f8" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "c3fe3d905bf555661f753798467885e1" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "6fb659e6272eba360035d57c318040df" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "8032e8c9a143a882ac2e010995e479f9" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "851e585dab37a4e334cd44c82a0746aa" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e3f7f20c0589afdf0eaede3356e51436" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "1233c34d1da2cc2ad765f4bffb1c2c14" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "e0d6bc78f25ab9a248d9973c527e25f1" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "97baace3417c96dccb06f2352b420dd4" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "7950614b6f04f9b2207da6e3ccbcf881" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "9689941523f41d428c6236542032a8f5" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "9ad73c8150d363228468c7ab61014a75" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "87311280f47b702cfccaf9bcdfdef592" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "1bb3c90399255004b0171def01d70799" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "55339b11830078c3ac94d1c104ed7fa0" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "d95b67a56bbc4683eae9324a2097f2e0" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "c6b396cb57055885ca075917d341c192" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "5730a76d1b9779ddfb2fdb9ad62cbbc7" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "5bd750b9118d52bc32f04da22d83d19a" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "522bdd82213944c3630d71029b7a6832" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "a62d14ef7fc04f6d702f98163f9f8d35" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "0767d7010664ac027e7c756e82060c67" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "c5ce67bb66d298c4cbeda17486a4d72e" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "b0ea40d107fd6eb8beb34274dc996c9f" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "9c993226c6613689f6a7063504c20b51" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "12096a370511b017a3acf02daf50ec0c" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "f3cfda2c63eea235840dc5f5d12c648e" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "3111ed1b68d9d7067e02cdbf7eaadb5c" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "0bd61912d2b5018bb81af2d82db61458" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e206538d20242a0810adfec972624c1e" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "ba5ac4629ff666cc3a5b36df3d6a2a6d" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "fa8abf418d3495a61a687c89f331d96c" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "9c857eafe781d624a4843651525aa408" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "bf17dd84f4d9c8c2fcb5ba17b8a778e7" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "744b52c129783cda67f8ff899cad1baf" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "4d846151427165ca49ac55bc96bca944" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7336e272e84afcbcbe9cfecb76e84a8b" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "9e115240167dc42e49d3ac108ab7685a" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8538d01854f867ed541f8b41449372af" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "a16eb68eec878be85b9bf52c2b2556d2" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "5882d414daab6c78acdb78cf22d3c1a4" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "3acd7eaa0f7d3d0aba4abb72bd141dcb" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "a60b315cbb0486bdedc46e41f314c7a8" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c5d63e3915864ada37e396e48391d0ac" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "88518a4e4ab3cfcc6d3c78100ea9a582" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ec30028bd1daeb278b8070a9aac6b374" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "0ac052b74b8006c2e28b518a1015b2e8" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "84a9bedde6c0e90397a0394a94617685" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "95b66dd9a7c21659a4cbdb8b09effdbb" + }, + { + "dataPath": "params_shard_114.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.28.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ee27bf11ec7c2ff3c2fe3d6a2a4a107e" + }, + { + "dataPath": "params_shard_115.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "fe956f4a536e6248b3e92087537c3363" + }, + { + "dataPath": "params_shard_116.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "25326f4b805daf13e2d35bad4ef5a3cb" + }, + { + "dataPath": "params_shard_117.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "3027ea34501e21fba6f9ed0c493d1074" + }, + { + "dataPath": "params_shard_118.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "a173a966625ee9b06007c9c536dbf8d0" + }, + { + "dataPath": "params_shard_119.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "7dbc95f1dc4b5a61481201b68f367e37" + }, + { + "dataPath": "params_shard_120.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "01f60a1470556669f44f087fc25f052a" + }, + { + "dataPath": "params_shard_121.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "4d3d309f4c6ab613b23afca60a2f1b5e" + }, + { + "dataPath": "params_shard_122.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.30.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "62662dfaae249b202ba8fb3a2759ad00" + }, + { + "dataPath": "params_shard_123.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3df05abab25c11bae55bf38fce144acc" + }, + { + "dataPath": "params_shard_124.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "35abab27c9339772c3fbb0a710320fa6" + }, + { + "dataPath": "params_shard_125.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "b1e633af5d88f9ecdbb0b946e4d0674a" + }, + { + "dataPath": "params_shard_126.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "c877a2561318b57ee02371012bd4a771" + }, + { + "dataPath": "params_shard_127.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "1ae0ef075d5a4aac5d223023dbcf82e0" + }, + { + "dataPath": "params_shard_128.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "e3b206f0b690a64e264fc6ea89ecb4ba" + }, + { + "dataPath": "params_shard_129.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.32.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "4ede5d81e1d8bb020b74aa0342a1e8b9" + }, + { + "dataPath": "params_shard_130.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.32.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "06700d6cbbc7b2e9464e3c291711c1fb" + }, + { + "dataPath": "params_shard_131.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.32.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "dcbdfb36013d002df3bf50b8f1984b50" + }, + { + "dataPath": "params_shard_132.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.32.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "05d809a37da496f1e62762787a4afe5d" + }, + { + "dataPath": "params_shard_133.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.33.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "059d417fad278cee9dfcf225ec2c1886" + }, + { + "dataPath": "params_shard_134.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.33.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "50f9a66c10fd4343455aef48f8f6c5b4" + }, + { + "dataPath": "params_shard_135.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.33.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "fbb8ae4aaf4433d91769f3ef6da52585" + }, + { + "dataPath": "params_shard_136.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.33.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "3ff0fa47e56ad4e2b7c5e171086b612b" + }, + { + "dataPath": "params_shard_137.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.34.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "48778061b8c2d61466ac449a9ba015ca" + }, + { + "dataPath": "params_shard_138.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.34.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "203b11e4d76b946a07d1d53a0d4f5109" + }, + { + "dataPath": "params_shard_139.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.34.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "c503e58aeb3755b0c846a2c884fcbcdb" + }, + { + "dataPath": "params_shard_140.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.34.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "52d20619a12643a1dc0cf815338cace4" + }, + { + "dataPath": "params_shard_141.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.35.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "343f22dca9d1ba8c86046427a125104c" + }, + { + "dataPath": "params_shard_142.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.35.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e9594bbf68f151d9d32ca7d086ed8180" + }, + { + "dataPath": "params_shard_143.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.35.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "5b0d0ffd3184319420ce6eb76beaafb5" + }, + { + "dataPath": "params_shard_144.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.35.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "afb305710472c81069655e5f2d567adb" + }, + { + "dataPath": "params_shard_145.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.36.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "0c442a11c5a9e973537b54f5e57db42e" + }, + { + "dataPath": "params_shard_146.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.36.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "b15dd5487e424d9fe1e8ca45c39b15a7" + }, + { + "dataPath": "params_shard_147.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.36.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "936a8af4c9e47844a18930b7bb12b4b5" + }, + { + "dataPath": "params_shard_148.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.36.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c65922d222392aea9351b73222564f1f" + }, + { + "dataPath": "params_shard_149.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.37.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "1418630cd93b77b170e0f06dd9770b9f" + }, + { + "dataPath": "params_shard_150.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.37.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "605ce68f42af16ab8c8a447df74f0f28" + }, + { + "dataPath": "params_shard_151.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.37.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "d035644d42d374e0a9da548292d9a8b2" + }, + { + "dataPath": "params_shard_152.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.37.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "01f3010ca5ac192baa5f827a198a4d0f" + }, + { + "dataPath": "params_shard_153.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.38.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "4bada577b216b41239e40e836dd11ef7" + }, + { + "dataPath": "params_shard_154.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.38.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "0a9b3fdd5b2b7cecb8db2472b6282bd3" + }, + { + "dataPath": "params_shard_155.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.38.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "0835173149ccea99084b1f98f37e69b6" + }, + { + "dataPath": "params_shard_156.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.38.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "1e644bebed08f6d9c215ecc570debe99" + }, + { + "dataPath": "params_shard_157.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.39.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7c24a06e0a8f18bb04809347faf36b31" + }, + { + "dataPath": "params_shard_158.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.39.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "0f90b3264cdac5af06ec79cbcf2b144a" + }, + { + "dataPath": "params_shard_159.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.39.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8c2e3f52b5b98eef9b423b10b667847a" + }, + { + "dataPath": "params_shard_160.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.39.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c31323c4d5ad3c5516c3cd66f4038e33" + }, + { + "dataPath": "params_shard_161.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.40.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "b67b08bdd7946d4ea51dde68c1898a38" + }, + { + "dataPath": "params_shard_162.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.40.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "b1d67c857e4e19bd9cac5f1b8868a119" + }, + { + "dataPath": "params_shard_163.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.40.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8a72d45cf2580e1a5b4122e4e493eeb7" + }, + { + "dataPath": "params_shard_164.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.40.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "bd4a6433ac721611548518ae4f508757" + }, + { + "dataPath": "params_shard_165.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.41.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "fea17431da3678196b33cbb3e7c4357f" + }, + { + "dataPath": "params_shard_166.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.41.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "5a188136524dd19a5ddf669933d8c887" + }, + { + "dataPath": "params_shard_167.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.41.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "6553eae95988d4c1282ecc1eca5ed86a" + }, + { + "dataPath": "params_shard_168.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.41.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "311a26abd3cad37da25b0f82ff5a4447" + }, + { + "dataPath": "params_shard_169.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.42.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "8b97995a2a142e78f76238b96ad4b979" + }, + { + "dataPath": "params_shard_170.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.42.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "99e3b250dcb189e7ac6635021d41ac60" + }, + { + "dataPath": "params_shard_171.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.42.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "077e76fa59e62d39baf7fc6df7fa6947" + }, + { + "dataPath": "params_shard_172.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.42.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "ef4fb5714a190a8ac7dc2b86ae5a1376" + }, + { + "dataPath": "params_shard_173.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.43.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "46505aac39d8a01ad918754a8039338e" + }, + { + "dataPath": "params_shard_174.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.43.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ebad9d410edefb2b9725937cfd0065d8" + }, + { + "dataPath": "params_shard_175.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.43.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "a0ad8234f679847ff20d0a525dbf171c" + }, + { + "dataPath": "params_shard_176.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.43.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "23d4a10eec0c0cef94c75074c5135424" + }, + { + "dataPath": "params_shard_177.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.44.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7e177f352900b047b91ae0bbf2568aa4" + }, + { + "dataPath": "params_shard_178.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.44.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ebb1ed234d8b3596646d57c1dc85a93f" + }, + { + "dataPath": "params_shard_179.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.44.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3b5adbcfec99f81286bf6d09a95bfbdc" + }, + { + "dataPath": "params_shard_180.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.44.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "b7446bdd156c62096f4d67f621b5d657" + }, + { + "dataPath": "params_shard_181.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.45.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "1ae707be88d06e9b3d019fc90e0b3703" + }, + { + "dataPath": "params_shard_182.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.45.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "2e3f59e7eb834c6b1bdcd61ec7386790" + }, + { + "dataPath": "params_shard_183.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.45.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "08285797dca9d40272f7795478c79e33" + }, + { + "dataPath": "params_shard_184.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.45.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "61aba3b03eba5c5646d0fdb96e508088" + }, + { + "dataPath": "params_shard_185.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.46.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "a578980050db5fc2121c7e7036d8579a" + }, + { + "dataPath": "params_shard_186.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.46.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "b11528a5ad8571babd007fe3b94e83d5" + }, + { + "dataPath": "params_shard_187.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.46.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "288c8f09e37324dabd6fdd7fcdb4d658" + }, + { + "dataPath": "params_shard_188.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.46.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "ceac600b994b51ffd539b62bfbfc1683" + }, + { + "dataPath": "params_shard_189.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.47.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "1d1853b03ea51ce5677d4934937457fc" + }, + { + "dataPath": "params_shard_190.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.47.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "46d3c2c29fc39614451bbb43e806ba53" + }, + { + "dataPath": "params_shard_191.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.47.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "28b517688ff920821b0f02133aedcf1a" + }, + { + "dataPath": "params_shard_192.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.47.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "94d361d73870fe14381eb763ef83e1aa" + }, + { + "dataPath": "params_shard_193.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.48.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "6de4af8747687645dc95ba25d8a6b88f" + }, + { + "dataPath": "params_shard_194.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.48.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ef80c903857f593d7d5a3e3f1fc80c3e" + }, + { + "dataPath": "params_shard_195.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.48.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "965497d8942e9df2b756482f2752412c" + }, + { + "dataPath": "params_shard_196.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.48.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "26a9be3702c28c1e0850295fd9b5abd4" + }, + { + "dataPath": "params_shard_197.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.49.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "a7ca94ca1acf52f8bf9ebdd61c85dfcb" + }, + { + "dataPath": "params_shard_198.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.49.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "44066d331e2604820f161a7fd62729aa" + }, + { + "dataPath": "params_shard_199.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.49.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "1770524da77c280da098facf2f4d8f17" + }, + { + "dataPath": "params_shard_200.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.49.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "b932117385b590732144624a16ca2eda" + }, + { + "dataPath": "params_shard_201.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.50.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "e92ba5a4c5390d85a30af79c5f4b2b9d" + }, + { + "dataPath": "params_shard_202.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.50.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "7fbb17bde368025353cc3bf246967aff" + }, + { + "dataPath": "params_shard_203.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.50.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "433af08a37bc3fb371ae2c759e1c5752" + }, + { + "dataPath": "params_shard_204.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.50.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "38243afc6b65c580b62ec94974236c19" + }, + { + "dataPath": "params_shard_205.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.51.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "4ca8917eb54e4b799f5f090c3894089a" + }, + { + "dataPath": "params_shard_206.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.51.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "70742ff3b5e0cff1d255fddb0092ff79" + }, + { + "dataPath": "params_shard_207.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.51.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "9bf7053f31761fc0e06aa8a1ad12a2e7" + }, + { + "dataPath": "params_shard_208.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.51.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "845720af2a4514d92c27978bd22c74d5" + }, + { + "dataPath": "params_shard_209.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.52.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "29f31e65158726cb39787463b21e835d" + }, + { + "dataPath": "params_shard_210.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.52.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "d24ad15583ff33c401bd637cd0e8794d" + }, + { + "dataPath": "params_shard_211.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.52.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "88e05091b81b82a058d901189ff003f4" + }, + { + "dataPath": "params_shard_212.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.52.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "6624bd17ba1572da0207971c8f408c93" + }, + { + "dataPath": "params_shard_213.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.53.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "b3fc317235240348e35785606534d809" + }, + { + "dataPath": "params_shard_214.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.53.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "29e4df9f7ce9e3d16d69837926c36638" + }, + { + "dataPath": "params_shard_215.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.53.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "be724864ed826b14266826f84caf0240" + }, + { + "dataPath": "params_shard_216.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.53.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "36cf05a535e939cec5e3befcfa69f49f" + }, + { + "dataPath": "params_shard_217.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.54.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "1150aa7c713a6c5d4926c3812a563266" + }, + { + "dataPath": "params_shard_218.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.54.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "f6ddfd24ca0de79bf9fc92d96039d1dd" + }, + { + "dataPath": "params_shard_219.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.54.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "8b17bcfa5ab585d136804f5f196404af" + }, + { + "dataPath": "params_shard_220.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.54.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c39d92db77bbaaf825747322b21c7809" + }, + { + "dataPath": "params_shard_221.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.55.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7f8e479d84a65852feed2d48536f7fa4" + }, + { + "dataPath": "params_shard_222.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.55.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "c007c060ca89b78a6eb16deff577b844" + }, + { + "dataPath": "params_shard_223.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.55.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "9a24df909bc31c8dcf23f124603404b1" + }, + { + "dataPath": "params_shard_224.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.55.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "8fd78bd7afb803b398a94f0f6ece9a3a" + }, + { + "dataPath": "params_shard_225.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.56.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "8ef84b18c389d39a2e8533c1f8c8e2df" + }, + { + "dataPath": "params_shard_226.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.56.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "744ec0099843a1f58bbfdcdcceb031c3" + }, + { + "dataPath": "params_shard_227.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.56.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "bb441888903596816bfcbb7998ebb252" + }, + { + "dataPath": "params_shard_228.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.56.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "27569e925413d2b4668281e64d98f799" + }, + { + "dataPath": "params_shard_229.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.57.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "6491f12abd3d0ae567077b812c954b36" + }, + { + "dataPath": "params_shard_230.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.57.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "ff4cbfe2815df1483bf37aa369cf5e90" + }, + { + "dataPath": "params_shard_231.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.57.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "d2a09e2e6da014066875b9c5e35aef73" + }, + { + "dataPath": "params_shard_232.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.57.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "3571f4f1c778575f8deb1748695c99c0" + }, + { + "dataPath": "params_shard_233.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.58.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "73efc8e8d056d75dc60a2bf55fb9c4dd" + }, + { + "dataPath": "params_shard_234.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.58.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "594100c2df0694d8ab3e78bf023f098b" + }, + { + "dataPath": "params_shard_235.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.58.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "2e414802788ca04fbf495999f7103b40" + }, + { + "dataPath": "params_shard_236.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.58.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "86f3107eed82246a0daca152666acce6" + }, + { + "dataPath": "params_shard_237.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.59.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "926c287122da6232754d04ecad75771e" + }, + { + "dataPath": "params_shard_238.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.59.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "bb4b4cfeea7ad616ad3a9967c8699293" + }, + { + "dataPath": "params_shard_239.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.59.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3fb105ad2bd22dbcbfe07e81c592d1af" + }, + { + "dataPath": "params_shard_240.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.59.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "92ed68713ae1a1de83d06fa386da5c8f" + }, + { + "dataPath": "params_shard_241.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.60.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "c9833719708bfa41e7513207ea445489" + }, + { + "dataPath": "params_shard_242.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.60.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e0462dbb6bde85209a9ef4ba8c5064cb" + }, + { + "dataPath": "params_shard_243.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.60.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "6f5cb3163973ab563f8a5bca1cbb0f0a" + }, + { + "dataPath": "params_shard_244.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.60.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "df48008d48cdea5a66c5d51208e1a0cc" + }, + { + "dataPath": "params_shard_245.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.61.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "d2d20696c35dd49899f34ecd4d4c65d2" + }, + { + "dataPath": "params_shard_246.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.61.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "3e0cdfa39f3ac71112712c9a24c9a113" + }, + { + "dataPath": "params_shard_247.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.61.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "692909387da52fcbe730dc7fe34cba9a" + }, + { + "dataPath": "params_shard_248.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.61.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "0ce22f04865a4109363a0ddef9a2f347" + }, + { + "dataPath": "params_shard_249.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.62.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "99b357a8f2494fcbfb54a1b19b44b1ed" + }, + { + "dataPath": "params_shard_250.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.62.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "645faeb07628beb438015c1c923c6f5d" + }, + { + "dataPath": "params_shard_251.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.62.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "66ef613612610ba683cefccb9ecfc5a7" + }, + { + "dataPath": "params_shard_252.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.62.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "db8b8ebaf42679625c49d7c9e2dde08a" + }, + { + "dataPath": "params_shard_253.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.63.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "90360acdc632b74ca526818f1d9c5d73" + }, + { + "dataPath": "params_shard_254.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.63.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "477e97877aef99ca8709fef7728137bd" + }, + { + "dataPath": "params_shard_255.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.63.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3089322f417d579bad053867844396a1" + }, + { + "dataPath": "params_shard_256.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.63.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "a2f7ff5b593db9318d6ddeacac55e57b" + }, + { + "dataPath": "params_shard_257.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.64.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "e7aa05329dbd063e5e77af4c2999d6fc" + }, + { + "dataPath": "params_shard_258.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.64.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "f1d148c6790c1a26af59e6be80002bb9" + }, + { + "dataPath": "params_shard_259.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.64.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3031239ee1d6b492358595292cc36a46" + }, + { + "dataPath": "params_shard_260.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.64.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "36febb44dd3652c6fd24e9052a2421c9" + }, + { + "dataPath": "params_shard_261.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.65.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "3605baafac3535492574f51d80a6047b" + }, + { + "dataPath": "params_shard_262.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.65.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "314abab4452c37e055183c42f48578db" + }, + { + "dataPath": "params_shard_263.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.65.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "dc5b30f0b0f434a8e1df9e4e43dfbe7c" + }, + { + "dataPath": "params_shard_264.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.65.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "449c6e77e3a4250eb30e1740e0f74d0c" + }, + { + "dataPath": "params_shard_265.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.66.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "3ced089ec744204585b81ca122e7c096" + }, + { + "dataPath": "params_shard_266.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.66.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "61c7b52295d4a9b9cbb04b34f519830c" + }, + { + "dataPath": "params_shard_267.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.66.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "17b50eeccfc55b174e4792320983d717" + }, + { + "dataPath": "params_shard_268.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.66.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "e57c0e788fb174dca4bcf8f6cc6ef3f9" + }, + { + "dataPath": "params_shard_269.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.67.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "01cfabcc143785ef7229fc6558896c2f" + }, + { + "dataPath": "params_shard_270.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.67.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "1b914f606656f5f0ccca9ca943178b2b" + }, + { + "dataPath": "params_shard_271.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.67.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "287cbfea7bea41daaa95414a378cfdd4" + }, + { + "dataPath": "params_shard_272.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.67.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "c7687124fddb949364f7b6116942245d" + }, + { + "dataPath": "params_shard_273.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.68.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "7461a27b1f19747ba13640645b3f9688" + }, + { + "dataPath": "params_shard_274.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.68.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "f3e29a300850a045b057c97c4aa24900" + }, + { + "dataPath": "params_shard_275.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.68.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "af01c00f3ca5689ea8458c0ab1d4c38f" + }, + { + "dataPath": "params_shard_276.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.68.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "1d3ff51e83424cca317027003ec6d246" + }, + { + "dataPath": "params_shard_277.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.69.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "6335d251e62a9ecdf4c8fe6da50f0334" + }, + { + "dataPath": "params_shard_278.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.69.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "3aedabf62b402ba20255b69a77148b95" + }, + { + "dataPath": "params_shard_279.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.69.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "b76db1e528f240172a98b2653697b97b" + }, + { + "dataPath": "params_shard_280.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.69.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "292a098561141c9afc24670ef468edc1" + }, + { + "dataPath": "params_shard_281.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.70.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "4d757b0cfe2bdab70353415bc08018ca" + }, + { + "dataPath": "params_shard_282.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.70.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "d1120ffea56a757690c36f698a78c0aa" + }, + { + "dataPath": "params_shard_283.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.70.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "77fb6a2a2040c0451b0ba2739acc3344" + }, + { + "dataPath": "params_shard_284.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.70.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "80a2c639a028ce12608f06cca1f91c68" + }, + { + "dataPath": "params_shard_285.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.71.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "283ea44c4313f668b5d8f9734695d050" + }, + { + "dataPath": "params_shard_286.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.71.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "61ca0cd25ffd8c4bcf47ca70aa9cd493" + }, + { + "dataPath": "params_shard_287.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.71.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "373499e281710a6ac79a9c109ef934f4" + }, + { + "dataPath": "params_shard_288.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.71.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "b1acef719570741f1ef6aa53d7bb3f4a" + }, + { + "dataPath": "params_shard_289.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.72.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "59b44c0fce2e66a65c6e0296b3283645" + }, + { + "dataPath": "params_shard_290.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.72.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "921f2e63992058cb5353265d047d5aa9" + }, + { + "dataPath": "params_shard_291.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.72.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "da190cbcb2731dcef328b0c2526efb62" + }, + { + "dataPath": "params_shard_292.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.72.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "f03dbbeb0d5a737b7015159c397ea17f" + }, + { + "dataPath": "params_shard_293.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.73.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "9e46044cf1b9174ddf88d2be5ba21a01" + }, + { + "dataPath": "params_shard_294.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.73.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "a432a29223cbe175881a3deb698334a6" + }, + { + "dataPath": "params_shard_295.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.73.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "bcc69e091f5a633c1b079fdd1d3790e1" + }, + { + "dataPath": "params_shard_296.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.73.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "410bc6fa7718724155ca258f62414909" + }, + { + "dataPath": "params_shard_297.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.74.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "dd0f8674cf9c4ffc9defe5d984468187" + }, + { + "dataPath": "params_shard_298.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.74.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "25f22b4ed224bc253483c2a4d4ecd4c8" + }, + { + "dataPath": "params_shard_299.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.74.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "17418ad25892c9116d4b92e4229948e9" + }, + { + "dataPath": "params_shard_300.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.74.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "413312f95783d87b1b254015b02fc9f2" + }, + { + "dataPath": "params_shard_301.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.75.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "ff6a14c311f77cccfbebfc0b0383ee1a" + }, + { + "dataPath": "params_shard_302.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.75.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "17c37f818bccce63b6ae0f59b3fa60bc" + }, + { + "dataPath": "params_shard_303.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.75.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "3fd72643daecdfda5f1ebf32442b565e" + }, + { + "dataPath": "params_shard_304.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.75.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "f7f13cd19786a7ced9d548cf40eda60e" + }, + { + "dataPath": "params_shard_305.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.76.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "dd230ee7e2190593add962a9a22576e1" + }, + { + "dataPath": "params_shard_306.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.76.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "e0e9850f981bd3c25c193249458140f9" + }, + { + "dataPath": "params_shard_307.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.76.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "96773422ec57df83679c4a1947c119cb" + }, + { + "dataPath": "params_shard_308.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.76.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "e64ccd59eeb4a66bfd55748b8d79bace" + }, + { + "dataPath": "params_shard_309.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.77.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "2683faa813b1f337047e919647679211" + }, + { + "dataPath": "params_shard_310.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.77.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "6966b3f34648ae4fff13d5905409425d" + }, + { + "dataPath": "params_shard_311.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.77.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "2f3370694ba3f6f210a3d7ccb980da07" + }, + { + "dataPath": "params_shard_312.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.77.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "8922603fc05b450b7ed6ddd0690d17ce" + }, + { + "dataPath": "params_shard_313.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.78.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "e096b3b9531ba75bb33ac9c0e545f605" + }, + { + "dataPath": "params_shard_314.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.78.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "643a8877b49a3600d630c0797ae12586" + }, + { + "dataPath": "params_shard_315.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.78.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "589e2d57067e275c0feb945c3feba8c3" + }, + { + "dataPath": "params_shard_316.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.78.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "d24dfd0dc9b6dbe637f7652609c46d48" + }, + { + "dataPath": "params_shard_317.bin", + "format": "raw-shard", + "nbytes": 167772160, + "records": [ + { + "name": "model.layers.79.self_attn.qkv_proj.weight", + "shape": [ + 10240, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 167772160, + "byteOffset": 0 + } + ], + "md5sum": "6c33a7c9dba7f110db40ea24847113d5" + }, + { + "dataPath": "params_shard_318.bin", + "format": "raw-shard", + "nbytes": 134217728, + "records": [ + { + "name": "model.layers.79.self_attn.o_proj.weight", + "shape": [ + 8192, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 134217728, + "byteOffset": 0 + } + ], + "md5sum": "853ffecb5f3e306a943c8874ca8238da" + }, + { + "dataPath": "params_shard_319.bin", + "format": "raw-shard", + "nbytes": 939524096, + "records": [ + { + "name": "model.layers.79.mlp.gate_up_proj.weight", + "shape": [ + 57344, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 939524096, + "byteOffset": 0 + } + ], + "md5sum": "d9b4a51d00e3d61a344b603609c39aff" + }, + { + "dataPath": "params_shard_320.bin", + "format": "raw-shard", + "nbytes": 469762048, + "records": [ + { + "name": "model.layers.79.mlp.down_proj.weight", + "shape": [ + 8192, + 28672 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 469762048, + "byteOffset": 0 + } + ], + "md5sum": "492cc10163c0936dfa6dd8bf5bf67eed" + }, + { + "dataPath": "params_shard_321.bin", + "format": "raw-shard", + "nbytes": 2101346304, + "records": [ + { + "name": "lm_head.weight", + "shape": [ + 128256, + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 2101346304, + "byteOffset": 0 + } + ], + "md5sum": "010c3ffb227e55714542e0722232c5c8" + }, + { + "dataPath": "params_shard_322.bin", + "format": "raw-shard", + "nbytes": 2637824, + "records": [ + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 0 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 16384 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 32768 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 49152 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 65536 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 81920 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 98304 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 114688 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 131072 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 147456 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 163840 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 180224 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 196608 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 212992 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 229376 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 245760 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 262144 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 278528 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 294912 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 311296 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 327680 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 344064 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 360448 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 376832 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 393216 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 409600 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 425984 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 442368 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 458752 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 475136 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 491520 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 507904 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 524288 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 540672 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 557056 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 573440 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 589824 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 606208 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 622592 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 638976 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 655360 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 671744 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 688128 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 704512 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 720896 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 737280 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 753664 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 770048 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 786432 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 802816 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 819200 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 835584 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 851968 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 868352 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 884736 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 901120 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 917504 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 933888 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 950272 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 966656 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 983040 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 999424 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1015808 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1032192 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1048576 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1064960 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1081344 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1097728 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1114112 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1130496 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1146880 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1163264 + }, + { + "name": "model.layers.36.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1179648 + }, + { + "name": "model.layers.36.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1196032 + }, + { + "name": "model.layers.37.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1212416 + }, + { + "name": "model.layers.37.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1228800 + }, + { + "name": "model.layers.38.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1245184 + }, + { + "name": "model.layers.38.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1261568 + }, + { + "name": "model.layers.39.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1277952 + }, + { + "name": "model.layers.39.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1294336 + }, + { + "name": "model.layers.40.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1310720 + }, + { + "name": "model.layers.40.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1327104 + }, + { + "name": "model.layers.41.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1343488 + }, + { + "name": "model.layers.41.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1359872 + }, + { + "name": "model.layers.42.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1376256 + }, + { + "name": "model.layers.42.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1392640 + }, + { + "name": "model.layers.43.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1409024 + }, + { + "name": "model.layers.43.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1425408 + }, + { + "name": "model.layers.44.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1441792 + }, + { + "name": "model.layers.44.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1458176 + }, + { + "name": "model.layers.45.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1474560 + }, + { + "name": "model.layers.45.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1490944 + }, + { + "name": "model.layers.46.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1507328 + }, + { + "name": "model.layers.46.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1523712 + }, + { + "name": "model.layers.47.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1540096 + }, + { + "name": "model.layers.47.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1556480 + }, + { + "name": "model.layers.48.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1572864 + }, + { + "name": "model.layers.48.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1589248 + }, + { + "name": "model.layers.49.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1605632 + }, + { + "name": "model.layers.49.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1622016 + }, + { + "name": "model.layers.50.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1638400 + }, + { + "name": "model.layers.50.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1654784 + }, + { + "name": "model.layers.51.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1671168 + }, + { + "name": "model.layers.51.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1687552 + }, + { + "name": "model.layers.52.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1703936 + }, + { + "name": "model.layers.52.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1720320 + }, + { + "name": "model.layers.53.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1736704 + }, + { + "name": "model.layers.53.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1753088 + }, + { + "name": "model.layers.54.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1769472 + }, + { + "name": "model.layers.54.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1785856 + }, + { + "name": "model.layers.55.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1802240 + }, + { + "name": "model.layers.55.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1818624 + }, + { + "name": "model.layers.56.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1835008 + }, + { + "name": "model.layers.56.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1851392 + }, + { + "name": "model.layers.57.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1867776 + }, + { + "name": "model.layers.57.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1884160 + }, + { + "name": "model.layers.58.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1900544 + }, + { + "name": "model.layers.58.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1916928 + }, + { + "name": "model.layers.59.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1933312 + }, + { + "name": "model.layers.59.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1949696 + }, + { + "name": "model.layers.60.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1966080 + }, + { + "name": "model.layers.60.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1982464 + }, + { + "name": "model.layers.61.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1998848 + }, + { + "name": "model.layers.61.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2015232 + }, + { + "name": "model.layers.62.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2031616 + }, + { + "name": "model.layers.62.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2048000 + }, + { + "name": "model.layers.63.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2064384 + }, + { + "name": "model.layers.63.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2080768 + }, + { + "name": "model.layers.64.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2097152 + }, + { + "name": "model.layers.64.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2113536 + }, + { + "name": "model.layers.65.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2129920 + }, + { + "name": "model.layers.65.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2146304 + }, + { + "name": "model.layers.66.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2162688 + }, + { + "name": "model.layers.66.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2179072 + }, + { + "name": "model.layers.67.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2195456 + }, + { + "name": "model.layers.67.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2211840 + }, + { + "name": "model.layers.68.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2228224 + }, + { + "name": "model.layers.68.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2244608 + }, + { + "name": "model.layers.69.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2260992 + }, + { + "name": "model.layers.69.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2277376 + }, + { + "name": "model.layers.70.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2293760 + }, + { + "name": "model.layers.70.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2310144 + }, + { + "name": "model.layers.71.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2326528 + }, + { + "name": "model.layers.71.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2342912 + }, + { + "name": "model.layers.72.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2359296 + }, + { + "name": "model.layers.72.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2375680 + }, + { + "name": "model.layers.73.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2392064 + }, + { + "name": "model.layers.73.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2408448 + }, + { + "name": "model.layers.74.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2424832 + }, + { + "name": "model.layers.74.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2441216 + }, + { + "name": "model.layers.75.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2457600 + }, + { + "name": "model.layers.75.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2473984 + }, + { + "name": "model.layers.76.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2490368 + }, + { + "name": "model.layers.76.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2506752 + }, + { + "name": "model.layers.77.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2523136 + }, + { + "name": "model.layers.77.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2539520 + }, + { + "name": "model.layers.78.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2555904 + }, + { + "name": "model.layers.78.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2572288 + }, + { + "name": "model.layers.79.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2588672 + }, + { + "name": "model.layers.79.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2605056 + }, + { + "name": "model.norm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2621440 + } + ], + "md5sum": "528c68d89451238af1af2a8e6e8eaab5" + } + ] +} \ No newline at end of file