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": 15.488057747664403 + }, + "records": [ + { + "dataPath": "params_shard_0.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": "e42ab2e294d24aa87ba90a2b2baaa1bb" + }, + { + "dataPath": "params_shard_1.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": "65ad07388d77341e98e31009fc421790" + }, + { + "dataPath": "params_shard_2.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": "f30f81ec7b5124fcba582de18c245c83" + }, + { + "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": "a0e6c528524565565b53cabf55ca483e" + }, + { + "dataPath": "params_shard_4.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": "1666c8d96e77810234a5f29b5ae54918" + }, + { + "dataPath": "params_shard_5.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": "67adcea9b2993963052cee4657046d68" + }, + { + "dataPath": "params_shard_6.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": "13801835040c6e45405955bece15ce39" + }, + { + "dataPath": "params_shard_7.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": "867110c0ba82f25cd0fc68583b1a1701" + }, + { + "dataPath": "params_shard_8.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": "9fc219ceebb2453bd6ccdad44153ade1" + }, + { + "dataPath": "params_shard_9.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": "480468beb1484d59a03d44205584c6ad" + }, + { + "dataPath": "params_shard_10.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": "4d8e8753557f24c6d79a959ff4468633" + }, + { + "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": "cf12b557a3bab2ad4de0577d4a9c7f7a" + }, + { + "dataPath": "params_shard_12.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": "7cfe93502082ce98beef6c2afe5b0611" + }, + { + "dataPath": "params_shard_13.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": "5bb17bc70d63a765bbe982595ed325f0" + }, + { + "dataPath": "params_shard_14.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": "74a6082f28effa1fe134e635934dda25" + }, + { + "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": "85af70da87baa38f98d9ccfa9b58b29e" + }, + { + "dataPath": "params_shard_16.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": "861d880fd8ee0b9d6220605b0909c345" + }, + { + "dataPath": "params_shard_17.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": "44873d8a70b4545ac73b2028cf637e25" + }, + { + "dataPath": "params_shard_18.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": "acf6c678e0cf36246933cf81ff05ecb4" + }, + { + "dataPath": "params_shard_19.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": "c397c7ae775d2e9b5db579bd6852b618" + }, + { + "dataPath": "params_shard_20.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": "659a8c27ceda4a327f2c76c8b9988980" + }, + { + "dataPath": "params_shard_21.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": "1259686ec8db673de3e60baf94a8a4e6" + }, + { + "dataPath": "params_shard_22.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": "8d4f882b044e97110b4e6dd88b776e38" + }, + { + "dataPath": "params_shard_23.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": "e26a3e3ba3b47b5d69de0e242de802c8" + }, + { + "dataPath": "params_shard_24.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": "239e8e3f2d144be3d799d407584a0004" + }, + { + "dataPath": "params_shard_25.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": "0c6c4f27f286ed93e8c20ac45d9ede9a" + }, + { + "dataPath": "params_shard_26.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": "9f7bdb7e4aff6c9c21137c31da5e4fc1" + }, + { + "dataPath": "params_shard_27.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": "9eff08cecb2fec081b88a83bc510f9a7" + }, + { + "dataPath": "params_shard_28.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": "a31aa746a326e97a52585f113eadaf64" + }, + { + "dataPath": "params_shard_29.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": "bc0ea9cae6e7fc8e48f4f9734e1487ce" + }, + { + "dataPath": "params_shard_30.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": "e42a08918f698672db209e063062ebfa" + }, + { + "dataPath": "params_shard_31.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": "2891758c66bed4520f224ac9f1c82af6" + }, + { + "dataPath": "params_shard_32.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": "15d1ee8f72a7899aa0dc550625828e30" + }, + { + "dataPath": "params_shard_33.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": "94eafc959c06e78afdaee8d9f4eb8d5a" + }, + { + "dataPath": "params_shard_34.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": "8bd6171d7a69b8575150d878e70bd58c" + }, + { + "dataPath": "params_shard_35.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": "3c18f0fdc1c4ca17c10bd757a25ab8d2" + }, + { + "dataPath": "params_shard_36.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": "105246a4d8e1307069ce3ce2df1847af" + }, + { + "dataPath": "params_shard_37.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": "653d3de1c98209d911646c145ca7f376" + }, + { + "dataPath": "params_shard_38.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": "9c3784b868bf76b4f6d6ae342043f385" + }, + { + "dataPath": "params_shard_39.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": "0e902d2c9f719e7b9de7bca013e1fd8a" + }, + { + "dataPath": "params_shard_40.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": "c0f4dbbb06f2892634bab78a7529f505" + }, + { + "dataPath": "params_shard_41.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": "d2c5df1dbd4981c34e755d1f37c25176" + }, + { + "dataPath": "params_shard_42.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": "deae312de6a2f595e493a18a7e6c1b83" + }, + { + "dataPath": "params_shard_43.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": "be6d1b56a8efd696614c463564c31320" + }, + { + "dataPath": "params_shard_44.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": "52f9f45894d1ebfffa9c582f7da25306" + }, + { + "dataPath": "params_shard_45.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": "db99046d9fff2259346c830de0d74caa" + }, + { + "dataPath": "params_shard_46.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": "d7e200da3586a8746991ed79378eb3de" + }, + { + "dataPath": "params_shard_47.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": "8fbfc9430ca205acb2cdd6eeba552cf9" + }, + { + "dataPath": "params_shard_48.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": "98bfd1c3befa52e571cf3f764f7b24db" + }, + { + "dataPath": "params_shard_49.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": "f94f6f84b0e9bbe16c26ba1100b58ab3" + }, + { + "dataPath": "params_shard_50.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": "628e8acbf0d421ac7653ad96f54008ce" + }, + { + "dataPath": "params_shard_51.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": "de71e3e3ac1f297684e4b750fe91886c" + }, + { + "dataPath": "params_shard_52.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": "32ed803eaa82fbabec32efb806d3949c" + }, + { + "dataPath": "params_shard_53.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": "2645a0ac4dc3c5a6b98fa2d9426be7a4" + }, + { + "dataPath": "params_shard_54.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": "cdffa481bb0b81617d3da5d4eee59ac8" + }, + { + "dataPath": "params_shard_55.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": "16590865f55096dfcac07fb8ab3f52c2" + }, + { + "dataPath": "params_shard_56.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": "a50451b0b33849bba9546665d4cd46c8" + }, + { + "dataPath": "params_shard_57.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": "d625b87c1828fbb5728d55e89b7b0fcf" + }, + { + "dataPath": "params_shard_58.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": "77f590dc0965f2488b260247b83d7e5b" + }, + { + "dataPath": "params_shard_59.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": "f670f6e7cb7ccb1ecd12c716a8d05132" + }, + { + "dataPath": "params_shard_60.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": "3b543d47793b25710f5b9aed19466de0" + }, + { + "dataPath": "params_shard_61.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": "1543421b96ef6044506795d72ac6faa2" + }, + { + "dataPath": "params_shard_62.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": "2545063fbc2a63ec2e62ff238f0550d7" + }, + { + "dataPath": "params_shard_63.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": "c454a9d3c58e614fcd9711de00c315fe" + }, + { + "dataPath": "params_shard_64.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": "031e6446649d867ee87b791ed1377e93" + }, + { + "dataPath": "params_shard_65.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": "2048f6fc285547a5e37bdc5115b8c2b7" + }, + { + "dataPath": "params_shard_66.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": "cff912beba70695b6d48c87b2096a4e5" + }, + { + "dataPath": "params_shard_67.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": "a515994781e1c268089dab25ebcdf2c5" + }, + { + "dataPath": "params_shard_68.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": "0fece08767d489cbd15ca8c7ac795799" + }, + { + "dataPath": "params_shard_69.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": "4aaf3993a17b47f5e4fbcbb56f456fde" + }, + { + "dataPath": "params_shard_70.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": "ce3c56a5e24cf4e2b30cc1250aaddd9f" + }, + { + "dataPath": "params_shard_71.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": "faa6e19d28c4391dd128a68c40acc88e" + }, + { + "dataPath": "params_shard_72.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": "1681aa6547840f2031dd56e9100ebfcf" + }, + { + "dataPath": "params_shard_73.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": "624b1a0a45fb8aa5f237a2df163f234c" + }, + { + "dataPath": "params_shard_74.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": "2f9e895501356905dbf15e61f5dd0e11" + }, + { + "dataPath": "params_shard_75.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": "92a150c6932de6dd42f33c574ed231c1" + }, + { + "dataPath": "params_shard_76.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": "f0534a9a1690ecb9c93dacef2f83b82a" + }, + { + "dataPath": "params_shard_77.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": "63a9457d07183668d6c5700c1f341b59" + }, + { + "dataPath": "params_shard_78.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": "87df162bfbc4d8c0e17337696d7f9dcc" + }, + { + "dataPath": "params_shard_79.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": "83455ae6ad325b0cdbb9b69cdeae98f3" + }, + { + "dataPath": "params_shard_80.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": "7a32ef1ab389c8c886fadd01786d68d7" + }, + { + "dataPath": "params_shard_81.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": "2d809dc42c4d9c18d674453a934baa9e" + }, + { + "dataPath": "params_shard_82.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": "7e32cf0f7290b0bdd147e4b9e0225651" + }, + { + "dataPath": "params_shard_83.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": "69a27253a15dd97d9f1f6fe84f463dd9" + }, + { + "dataPath": "params_shard_84.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": "5fa5031fac7fdfb5e269257618f0db9b" + }, + { + "dataPath": "params_shard_85.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": "43f342a97e52f64b08d156d1387c2edf" + }, + { + "dataPath": "params_shard_86.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": "0ca8336698c951cbf80db1cb4cbf4cc7" + }, + { + "dataPath": "params_shard_87.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": "f5b1a68d21384d85b188368a9789ac37" + }, + { + "dataPath": "params_shard_88.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": "d64dd2125cdf08941f20a37b60daadd8" + }, + { + "dataPath": "params_shard_89.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": "765932ab2279cf35f25fadf6d3bdb04b" + }, + { + "dataPath": "params_shard_90.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": "439e86603b7f1037553b169e7dad55fb" + }, + { + "dataPath": "params_shard_91.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": "7450a6a9326a79f105cd2fc131517f6f" + }, + { + "dataPath": "params_shard_92.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": "017511d4d8427deb1dd204a3910a0b78" + }, + { + "dataPath": "params_shard_93.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": "d9d3f2c61d4088baf7763658a718e8ee" + }, + { + "dataPath": "params_shard_94.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": "bacbf4aada39b0a26215998e9875386d" + }, + { + "dataPath": "params_shard_95.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": "408b308273476fac8487dd1cfa65d92c" + }, + { + "dataPath": "params_shard_96.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": "7aeb9337520011ce73edff58a8d149d2" + }, + { + "dataPath": "params_shard_97.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": "4e678d08f10a4a4cc65955a9186751d8" + }, + { + "dataPath": "params_shard_98.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": "d4ae7a0b46598484b419aa6898de67ec" + }, + { + "dataPath": "params_shard_99.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": "b295d6ed07c97752d915d2e40e1ac1f6" + }, + { + "dataPath": "params_shard_100.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": "7c1041e8c4caa8464a08b1563354aeea" + }, + { + "dataPath": "params_shard_101.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": "1b15792b7875b687ce5324f2a547bd7b" + }, + { + "dataPath": "params_shard_102.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": "e28cf8c68c76cf45b5d6bce0ccc23374" + }, + { + "dataPath": "params_shard_103.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": "d52cdbee829cb66a7ab382d1e3445f96" + }, + { + "dataPath": "params_shard_104.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": "ad3e007b8777a8538086aeef7a50e113" + }, + { + "dataPath": "params_shard_105.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": "51c966d60453008f3f9dde01d7f4717f" + }, + { + "dataPath": "params_shard_106.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": "7f25f96e95c59fb609326ea1440fdfdb" + }, + { + "dataPath": "params_shard_107.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": "b94a11124080ae42114fe4b085e505c4" + }, + { + "dataPath": "params_shard_108.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": "3930552116b7be6b9cf38388d7e079ad" + }, + { + "dataPath": "params_shard_109.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": "88e7121fa3db618ac96ccd10d4eea10a" + }, + { + "dataPath": "params_shard_110.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": "14485a2c5f44526f1e710caf16a2e61b" + }, + { + "dataPath": "params_shard_111.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": "a92bbeaa52a54f59955d5bf643e104ba" + }, + { + "dataPath": "params_shard_112.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": "e48bfd25323d041ed51ce42a6b7adb77" + }, + { + "dataPath": "params_shard_113.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": "897aafc8f4dc8d4ecfd7d0026e445d8c" + }, + { + "dataPath": "params_shard_114.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": "da9bc48cfab3327dcafae84116f932e2" + }, + { + "dataPath": "params_shard_115.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": "46bbca1f968d72b19ca5008e45f730e3" + }, + { + "dataPath": "params_shard_116.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": "a40549b63b3c02a90edd1eaddd998d13" + }, + { + "dataPath": "params_shard_117.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": "a9a3285ed58d399a0dbb2d6201f2dd6c" + }, + { + "dataPath": "params_shard_118.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": "bff04f3672ce0f4b0e33aba569cd2f38" + }, + { + "dataPath": "params_shard_119.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": "13cf3658181f5538bd3f3d1780bc156b" + }, + { + "dataPath": "params_shard_120.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": "1af43cd544e941eba9c20b572f1b301a" + }, + { + "dataPath": "params_shard_121.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": "d4f4bd0a55b61ac574b13db2e976798c" + }, + { + "dataPath": "params_shard_122.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": "733fd8714b0b52a7aef873e649d56a98" + }, + { + "dataPath": "params_shard_123.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": "074043f4c8b527249340b864658490ed" + }, + { + "dataPath": "params_shard_124.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": "ab920eae3bf19d72336cfbbb3e7bfe9b" + }, + { + "dataPath": "params_shard_125.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": "f003f0c9bc713f94e005b8a27d10c712" + }, + { + "dataPath": "params_shard_126.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": "56422cff615446b4fd7a4a45f44d0858" + }, + { + "dataPath": "params_shard_127.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": "467d1f430b9430f563e8a0f55a0cba0c" + }, + { + "dataPath": "params_shard_128.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": "084b05ea4e92fd8e121b71cac22e2259" + }, + { + "dataPath": "params_shard_129.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": "e66577da33ab718173804fd529893d57" + }, + { + "dataPath": "params_shard_130.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": "2546f2fff6d3c994847d67126314c085" + }, + { + "dataPath": "params_shard_131.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": "aa2350e5a4f5231ccb83b2ca9d9c4c22" + }, + { + "dataPath": "params_shard_132.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": "86411163aeaa57d055f4a9ed1d3cb9d0" + }, + { + "dataPath": "params_shard_133.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": "e793fa2c49cbc47880e32c78d2c06ffa" + }, + { + "dataPath": "params_shard_134.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": "61a2b76a0e7055669a15bcf0e33e068f" + }, + { + "dataPath": "params_shard_135.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": "7af5a5d36ed5e7e215e026c580cbf0b0" + }, + { + "dataPath": "params_shard_136.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": "578e499a0f3bc5cdef3b8957e56ff15d" + }, + { + "dataPath": "params_shard_137.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": "cf5533348f6b588e7049a045cf49839b" + }, + { + "dataPath": "params_shard_138.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": "6401ba08c1ce942191c8f8831a77f0fd" + }, + { + "dataPath": "params_shard_139.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": "e786763b4783d3359bc3ec8e11ac5e88" + }, + { + "dataPath": "params_shard_140.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": "da7b14ffa9580e595d9b96005a1be7c4" + }, + { + "dataPath": "params_shard_141.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": "9ba7abfaa462c5f72103c0d22efdb8cd" + }, + { + "dataPath": "params_shard_142.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": "8cbbe219ec056caed2a02f44c77e0086" + }, + { + "dataPath": "params_shard_143.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": "a97a9afa6f10b7a5e87522dbca21b4e2" + }, + { + "dataPath": "params_shard_144.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": "057b3e8173b74751370d3d2b242e88d9" + }, + { + "dataPath": "params_shard_145.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": "3fc54f818bc0aed8cd5208443a658899" + }, + { + "dataPath": "params_shard_146.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": "3497cc93ae7d342ba62f7de06cba98b6" + }, + { + "dataPath": "params_shard_147.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": "2090bdcdd46ef8cc24feb01b9cb0a00c" + }, + { + "dataPath": "params_shard_148.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": "dcb063fe18791b0807ae8fbea1094538" + }, + { + "dataPath": "params_shard_149.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": "1a6f4609352e7527020f3762b04da414" + }, + { + "dataPath": "params_shard_150.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": "9e5152d4a8f69c718f69b460fc59af7e" + }, + { + "dataPath": "params_shard_151.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": "50fcb738ddf73453dde9640067a9782c" + }, + { + "dataPath": "params_shard_152.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": "c960d7ad770d5769981d9ee7d4c9f94c" + }, + { + "dataPath": "params_shard_153.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": "a57ac220d92aaff6dec62216fd6cb3aa" + }, + { + "dataPath": "params_shard_154.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": "c459233a11cf9889ad6c2b592b01e9c7" + }, + { + "dataPath": "params_shard_155.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": "b00a8c9ddf83820e54019f48ff1f64c7" + }, + { + "dataPath": "params_shard_156.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": "e7630d8d7675088d0c0f59fadcb05f39" + }, + { + "dataPath": "params_shard_157.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": "c2765f9fa299ded2996ede8b8ad174b8" + }, + { + "dataPath": "params_shard_158.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": "480916d683ff6dd95160f990582e8e7f" + }, + { + "dataPath": "params_shard_159.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": "15f7ab9f0ef2cff30fe3b6e932e55789" + }, + { + "dataPath": "params_shard_160.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": "19518be8f82e064f65c3e5f9c2348082" + }, + { + "dataPath": "params_shard_161.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": "0c650be2cd87e3b17c98a24718dd8314" + }, + { + "dataPath": "params_shard_162.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": "f20fb3762e79fcee35474804ad4e95ef" + }, + { + "dataPath": "params_shard_163.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": "370a63688246772966c2cd912c23e6cb" + }, + { + "dataPath": "params_shard_164.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": "bcda1af1383385a463ba7cafb6debdc2" + }, + { + "dataPath": "params_shard_165.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": "780aa561474317349003dd9d59792204" + }, + { + "dataPath": "params_shard_166.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": "2e11eb819ebaaa5ef8f809861225a66b" + }, + { + "dataPath": "params_shard_167.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": "fc88a556aaf585aee311a7140baf12d0" + }, + { + "dataPath": "params_shard_168.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": "bea3efdfa32275b142d99b33b831c4fe" + }, + { + "dataPath": "params_shard_169.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": "603e36f0e3e68656fb852c25c6fafb2c" + }, + { + "dataPath": "params_shard_170.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": "7b7fc866680168f7baa56c95080cd605" + }, + { + "dataPath": "params_shard_171.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": "a22b025c08ce38a6aee40384784f9495" + }, + { + "dataPath": "params_shard_172.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": "919dcf897df2cce67354912ea951df8e" + }, + { + "dataPath": "params_shard_173.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": "3ca327330c6a75c372c7c15f2b252822" + }, + { + "dataPath": "params_shard_174.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": "3af36056f8b8f71003a6633110eca12c" + }, + { + "dataPath": "params_shard_175.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": "a9cd81d4d73996fd8613e80965424cf1" + }, + { + "dataPath": "params_shard_176.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": "76485173acac81702f89ca40bd997943" + }, + { + "dataPath": "params_shard_177.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": "99f1ed6ec6cb9f6542e8908fb6373331" + }, + { + "dataPath": "params_shard_178.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": "09787924e6019ed5e44037ab19d985e9" + }, + { + "dataPath": "params_shard_179.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": "fdc8a930ddb76fde1399b050b29e98f6" + }, + { + "dataPath": "params_shard_180.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": "02bd6a14991afe2aa6e8580f8f338a9c" + }, + { + "dataPath": "params_shard_181.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": "3986f7e89373c6f81aa45f42cc7760ab" + }, + { + "dataPath": "params_shard_182.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": "ae63850a704a3b9c7ec83857c52fb0d6" + }, + { + "dataPath": "params_shard_183.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": "72982bcbda1f6784af360e4c3a81aee5" + }, + { + "dataPath": "params_shard_184.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": "5aba0e1a61edc43af5d1793da26700b3" + }, + { + "dataPath": "params_shard_185.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": "b7e6ac6b711940814e181ab5e5fee40e" + }, + { + "dataPath": "params_shard_186.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": "2139ab3918f6f5203516421b8740f698" + }, + { + "dataPath": "params_shard_187.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": "4fd23b41c548afdcbda83e4443bcbde7" + }, + { + "dataPath": "params_shard_188.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": "dc787b6e3b767753b09c5ec352181a08" + }, + { + "dataPath": "params_shard_189.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": "a0c9a040625a906f8774559792d4a2d0" + }, + { + "dataPath": "params_shard_190.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": "03fe5875d7c00196aaa25cf4b1b97db0" + }, + { + "dataPath": "params_shard_191.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": "c74a7bf9debdbe5cff8e3e45ddc634a5" + }, + { + "dataPath": "params_shard_192.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": "8461a73a3608a7a317f1d06a944876bc" + }, + { + "dataPath": "params_shard_193.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": "1783ea73a7d40fedddc2ae333e9f6b4d" + }, + { + "dataPath": "params_shard_194.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": "dd3a8d1d3237a67bdc26650f0a876159" + }, + { + "dataPath": "params_shard_195.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": "d05077dffbd704152254135cb141cc79" + }, + { + "dataPath": "params_shard_196.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": "9686124280afdbc465ac7616fe7cacce" + }, + { + "dataPath": "params_shard_197.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": "b39f6ddd94c844a080d177b00b24e180" + }, + { + "dataPath": "params_shard_198.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": "127bddeff2e39659b18cd832529153ff" + }, + { + "dataPath": "params_shard_199.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": "e59eaf2bd794f13f4cb3251c7fe35e85" + }, + { + "dataPath": "params_shard_200.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": "b526d29c36bc7e9b60f9e88186a6a388" + }, + { + "dataPath": "params_shard_201.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": "35c6be97a34eb5724c04718e3f9582e3" + }, + { + "dataPath": "params_shard_202.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": "af7459103fe6a0d212b5278088d7f2e3" + }, + { + "dataPath": "params_shard_203.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": "5b31ed4b443f4de99a3bdeadfcd5a890" + }, + { + "dataPath": "params_shard_204.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": "31cc602a5ee3b3c3cc806f663fce556b" + }, + { + "dataPath": "params_shard_205.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": "f70ac4b630a856858e3ca87d47234718" + }, + { + "dataPath": "params_shard_206.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": "82d8d36a41f330f1c8e5997ffe21d22f" + }, + { + "dataPath": "params_shard_207.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": "9d338900a4408fd0c94ff4194bfb2fa8" + }, + { + "dataPath": "params_shard_208.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": "ce884c8b69c530088e91a25c3967a794" + }, + { + "dataPath": "params_shard_209.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": "fd7f9867e48a379e50e1bd138b130edd" + }, + { + "dataPath": "params_shard_210.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": "5c3a03bdc944af2551cdb9d9e8f1fc43" + }, + { + "dataPath": "params_shard_211.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": "9a4a69466aa43d1fcdde7a2eedcc4613" + }, + { + "dataPath": "params_shard_212.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": "77a923b38e61aece52201e2915ae8642" + }, + { + "dataPath": "params_shard_213.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": "94a0a6ed574e66d8370bfae1e1138a52" + }, + { + "dataPath": "params_shard_214.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": "b76434d4a2d565196efac188d9d9ec6a" + }, + { + "dataPath": "params_shard_215.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": "a7a476fbaf303c7ed3c9937456cc31d3" + }, + { + "dataPath": "params_shard_216.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": "5e9a202cdf52733eb2f6232244644013" + }, + { + "dataPath": "params_shard_217.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": "6466619703ed342e84f4e20bec4d6ed5" + }, + { + "dataPath": "params_shard_218.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": "9ba4fb25b036a0b83ce3338fa7db86fb" + }, + { + "dataPath": "params_shard_219.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": "0ec672202aaf3ba2de8a2b0dd628a37c" + }, + { + "dataPath": "params_shard_220.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": "217011736a023bcafab5bb332cc8beac" + }, + { + "dataPath": "params_shard_221.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": "c12e3ed1ad8231f63983ce856fb79c5b" + }, + { + "dataPath": "params_shard_222.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": "acf1f1f075197dadbb59250bc28ddf22" + }, + { + "dataPath": "params_shard_223.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": "126560528abe372eaf58ba525bac94de" + }, + { + "dataPath": "params_shard_224.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": "a1207c87b95faa592809a66ea31547af" + }, + { + "dataPath": "params_shard_225.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": "88a60562508bc4ad28551ea8ed0fbb42" + }, + { + "dataPath": "params_shard_226.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": "922ce5852ad654550597f4dcd06bd56a" + }, + { + "dataPath": "params_shard_227.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": "dfbc51701c2af3b948161e112be9af7b" + }, + { + "dataPath": "params_shard_228.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": "30b49cb0f0b54b9af3d734cef32aa617" + }, + { + "dataPath": "params_shard_229.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": "68a5f11216db6e17f0aca72caa36dc56" + }, + { + "dataPath": "params_shard_230.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": "7e34a8f43a1c53829fafb35096be93cf" + }, + { + "dataPath": "params_shard_231.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": "9407cc63589a3fbb6f631859524d7e6e" + }, + { + "dataPath": "params_shard_232.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": "df506579b007ad84efd29daa9864a1a2" + }, + { + "dataPath": "params_shard_233.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": "b51002a4c1989e4358574b4b9d3b5f87" + }, + { + "dataPath": "params_shard_234.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": "4f345098ca7cfe0ae6538f035a3ccbe4" + }, + { + "dataPath": "params_shard_235.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": "e047f498215be06be1f76f34e5e34fde" + }, + { + "dataPath": "params_shard_236.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": "91269b1bfaf33e5702be8476c294c6cc" + }, + { + "dataPath": "params_shard_237.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": "0d79852d4600cca1c279d4a2bc435c8a" + }, + { + "dataPath": "params_shard_238.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": "f996d0d1fa4bb5166dd42221d0c39480" + }, + { + "dataPath": "params_shard_239.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": "e42684535f2ae7319ba6c7a7e2c7c58a" + }, + { + "dataPath": "params_shard_240.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": "02ce0591180c65fe1f9aa7a6e9546fff" + }, + { + "dataPath": "params_shard_241.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": "c8c77b6a82c166eaa1b9d29f3056d761" + }, + { + "dataPath": "params_shard_242.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": "ecfa6ee66f62cdaf369223d7f895eb0b" + }, + { + "dataPath": "params_shard_243.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": "93991cbc7785d1534377f4a576f61456" + }, + { + "dataPath": "params_shard_244.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": "5828e6a47aeebe885110f923efdd3822" + }, + { + "dataPath": "params_shard_245.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": "fe9c854841d4b4963ce3fe277a4e2803" + }, + { + "dataPath": "params_shard_246.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": "d560ded8942eae3b9c128927837896bc" + }, + { + "dataPath": "params_shard_247.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": "26bf27814e9590bbdd4c1c75677e72ad" + }, + { + "dataPath": "params_shard_248.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": "ef11f5a0e2d4059ca05b95caa9519aa6" + }, + { + "dataPath": "params_shard_249.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": "72f02083032b90d7b84c08485c002270" + }, + { + "dataPath": "params_shard_250.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": "233e4d20ccb93c00e9e5aa6d2e4e110f" + }, + { + "dataPath": "params_shard_251.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": "d9764ba8b79e12c8b5e6eb4adb5fbc21" + }, + { + "dataPath": "params_shard_252.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": "443982329034fc264ec70337bb7d6994" + }, + { + "dataPath": "params_shard_253.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": "74f91752694002cccbf6d275ea7622f5" + }, + { + "dataPath": "params_shard_254.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": "d6f473fb5be03c9df59b2442a2a95a31" + }, + { + "dataPath": "params_shard_255.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": "09b98f1de9b2231c6bdc26b6581ce872" + }, + { + "dataPath": "params_shard_256.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": "621c3d9fb2cdbfd37a8c01b73d88c039" + }, + { + "dataPath": "params_shard_257.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": "21dcdfa4ad91d2cf22778a51469c51f4" + }, + { + "dataPath": "params_shard_258.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": "49bedfa5ecbfd20ff5c1340aa68ecd89" + }, + { + "dataPath": "params_shard_259.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": "78e24f3ecb46f3d961945fce95cc734e" + }, + { + "dataPath": "params_shard_260.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": "c800ef548ed9326e872c211c0b5a3db1" + }, + { + "dataPath": "params_shard_261.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": "02ca9d62ef139a607325d30a3d52b529" + }, + { + "dataPath": "params_shard_262.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": "17e19bd282c57a24ab9914036e416c09" + }, + { + "dataPath": "params_shard_263.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": "f7adf9c31ea41d1d743e26787332b089" + }, + { + "dataPath": "params_shard_264.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": "a4fc9349e06ef31ccdfcf20ab60a1758" + }, + { + "dataPath": "params_shard_265.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": "ccc852fea4d589f10a20d6f6c7f63c49" + }, + { + "dataPath": "params_shard_266.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": "a31bd2051a6a72332647e26f63c396bf" + }, + { + "dataPath": "params_shard_267.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": "1e61288165fdb578482aa5a5ac0ed1f4" + }, + { + "dataPath": "params_shard_268.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": "e21eb865107c3ec4da7c1b5d682d55b1" + }, + { + "dataPath": "params_shard_269.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": "80a6db1633fc0f320e91588a9cc914dc" + }, + { + "dataPath": "params_shard_270.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": "a7cb307c01e36a792777073dfa28953f" + }, + { + "dataPath": "params_shard_271.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": "03ab09af4c2a1934f4390d34cb979119" + }, + { + "dataPath": "params_shard_272.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": "c0e111c36f99c0e8cd2f89bcd4ce7f78" + }, + { + "dataPath": "params_shard_273.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": "58dc7bb165b13ac35030085f2e2c8275" + }, + { + "dataPath": "params_shard_274.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": "115f3c372907d20592e3cbd1ab108cf1" + }, + { + "dataPath": "params_shard_275.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": "b442d66a34b754e5bf9d9d1088f69d51" + }, + { + "dataPath": "params_shard_276.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": "e794189a9ae11852d2cd52452f458816" + }, + { + "dataPath": "params_shard_277.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": "e4687ed44ad0ca8ba867a8e5e1d3cd0d" + }, + { + "dataPath": "params_shard_278.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": "8b85c2c6c4f057d641b2c3e491f19a11" + }, + { + "dataPath": "params_shard_279.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": "ebcee6dddf9d9890ef4ee40daf948dff" + }, + { + "dataPath": "params_shard_280.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": "d61343108c6fca22deedb0fb539ad604" + }, + { + "dataPath": "params_shard_281.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": "7da0de226695601d6612df2ea8d82971" + }, + { + "dataPath": "params_shard_282.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": "768313c8dc22c9552460490e8581f628" + }, + { + "dataPath": "params_shard_283.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": "d2804c6e1db3a86f1e9f66b1c6d68c95" + }, + { + "dataPath": "params_shard_284.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": "84ad7ebed4bbd34599e5e03f69a29b1f" + }, + { + "dataPath": "params_shard_285.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": "104551086d1944d21eb8012f007f0e6b" + }, + { + "dataPath": "params_shard_286.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": "cb38084027ae1d8c9e7a978817fc1495" + }, + { + "dataPath": "params_shard_287.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": "9c5b47873f3ea6f560b49be976b3d90b" + }, + { + "dataPath": "params_shard_288.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": "58a6c29293bf136e769b278f43eca07d" + }, + { + "dataPath": "params_shard_289.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": "020f1cc094fc2a4f065a3ab4c2cf5a8a" + }, + { + "dataPath": "params_shard_290.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": "2a614c7b8cb869f8e7d33a358ca83b11" + }, + { + "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": "7e65d4644d4cee87af9217b6e62b1565" + }, + { + "dataPath": "params_shard_292.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": "dd81ec355ea9d1dfd8762c47e0e1c0aa" + }, + { + "dataPath": "params_shard_293.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": "8f80e7a1a9b0a062cbba46321a01137e" + }, + { + "dataPath": "params_shard_294.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": "ce20f6bd87fff8daab83b114fc8a313a" + }, + { + "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": "9dd5cb258a514b566cfc94955477fc56" + }, + { + "dataPath": "params_shard_296.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": "67972d93fcee98b41ed33c8ecf34df0d" + }, + { + "dataPath": "params_shard_297.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": "dc3f23b717580785278151e0b9f32956" + }, + { + "dataPath": "params_shard_298.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": "d1da22e9baaacd684ebffb0e9abbdfba" + }, + { + "dataPath": "params_shard_299.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": "8cd0417abec9ff9194a0c3105673968e" + }, + { + "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": "4bf68f9d373712380b35b5cb3cc1f481" + }, + { + "dataPath": "params_shard_301.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": "f32c7878ab91993b6e0d4b8dce39fd9e" + }, + { + "dataPath": "params_shard_302.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": "0d9d0291e6f7a12642b10bdd3c2a86bb" + }, + { + "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": "ecba900c78ab1a833815e20943dd1150" + }, + { + "dataPath": "params_shard_304.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": "89e66a7e2dea8cd0a68dd82e7b804a91" + }, + { + "dataPath": "params_shard_305.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": "2e4550abd46acd4bf8b72541c5225eeb" + }, + { + "dataPath": "params_shard_306.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": "e3b41b334d6eb1b0fd102b535d03a868" + }, + { + "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": "b7c84ba4420051467baa2b5cfd64feab" + }, + { + "dataPath": "params_shard_308.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": "0d1e3ae2f38396fac185cd2f6f5f6a46" + }, + { + "dataPath": "params_shard_309.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": "7338d19ac26286fd088e68a72411023c" + }, + { + "dataPath": "params_shard_310.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": "3880b574182ef74f5e3001696cc6e9f9" + }, + { + "dataPath": "params_shard_311.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": "a881a499e8fc3662432ecb9bad520d89" + }, + { + "dataPath": "params_shard_312.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": "f2cd5d0542af10fef9f3c1bdb98c4476" + }, + { + "dataPath": "params_shard_313.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": "a24b760b2961a925ba607eb28857173e" + }, + { + "dataPath": "params_shard_314.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": "ef718c8ff790f8426b9a87f15779d302" + }, + { + "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": "15ddaca82498fd442906b6eca0e919de" + }, + { + "dataPath": "params_shard_316.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": "115a7e758709fee1a9d6a99c98af2395" + }, + { + "dataPath": "params_shard_317.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": "016a6c5f3373c2fd8568bab32f5503a2" + }, + { + "dataPath": "params_shard_318.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": "542fb35deb3f8ab4837dad4143cb02bf" + }, + { + "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": "c2f57db509de34cc42f1ad6ca18cb3e0" + }, + { + "dataPath": "params_shard_320.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": "1dc20c2c457af7416cef56e5f443760a" + }, + { + "dataPath": "params_shard_321.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": "ee84a5ab6ec01f89657909008ad9926a" + }, + { + "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.10.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 131072 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 147456 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 163840 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 180224 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 196608 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 212992 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 229376 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 245760 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 262144 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 278528 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 294912 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 311296 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 327680 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 344064 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 360448 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 376832 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 393216 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 409600 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 425984 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 442368 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 458752 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 475136 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 491520 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 507904 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 524288 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 540672 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 557056 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 573440 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 589824 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 606208 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 622592 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 638976 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 655360 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 671744 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 688128 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 704512 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 720896 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 737280 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 753664 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 770048 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 786432 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 802816 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 819200 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 835584 + }, + { + "name": "model.layers.32.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 851968 + }, + { + "name": "model.layers.32.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 868352 + }, + { + "name": "model.layers.33.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 884736 + }, + { + "name": "model.layers.33.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 901120 + }, + { + "name": "model.layers.34.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 917504 + }, + { + "name": "model.layers.34.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 933888 + }, + { + "name": "model.layers.35.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 950272 + }, + { + "name": "model.layers.35.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 966656 + }, + { + "name": "model.layers.36.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 983040 + }, + { + "name": "model.layers.36.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 999424 + }, + { + "name": "model.layers.37.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1015808 + }, + { + "name": "model.layers.37.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1032192 + }, + { + "name": "model.layers.38.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1048576 + }, + { + "name": "model.layers.38.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1064960 + }, + { + "name": "model.layers.39.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1081344 + }, + { + "name": "model.layers.39.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1097728 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1114112 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1130496 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1146880 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1163264 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1179648 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1196032 + }, + { + "name": "model.layers.40.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1212416 + }, + { + "name": "model.layers.40.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1228800 + }, + { + "name": "model.layers.41.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1245184 + }, + { + "name": "model.layers.41.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1261568 + }, + { + "name": "model.layers.42.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1277952 + }, + { + "name": "model.layers.42.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1294336 + }, + { + "name": "model.layers.43.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1310720 + }, + { + "name": "model.layers.43.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1327104 + }, + { + "name": "model.layers.44.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1343488 + }, + { + "name": "model.layers.44.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1359872 + }, + { + "name": "model.layers.45.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1376256 + }, + { + "name": "model.layers.45.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1392640 + }, + { + "name": "model.layers.46.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1409024 + }, + { + "name": "model.layers.46.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1425408 + }, + { + "name": "model.layers.47.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1441792 + }, + { + "name": "model.layers.47.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1458176 + }, + { + "name": "model.layers.48.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1474560 + }, + { + "name": "model.layers.48.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1490944 + }, + { + "name": "model.layers.49.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1507328 + }, + { + "name": "model.layers.49.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1523712 + }, + { + "name": "model.layers.50.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1540096 + }, + { + "name": "model.layers.50.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1556480 + }, + { + "name": "model.layers.51.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1572864 + }, + { + "name": "model.layers.51.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1589248 + }, + { + "name": "model.layers.52.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1605632 + }, + { + "name": "model.layers.52.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1622016 + }, + { + "name": "model.layers.53.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1638400 + }, + { + "name": "model.layers.53.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1654784 + }, + { + "name": "model.layers.54.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1671168 + }, + { + "name": "model.layers.54.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1687552 + }, + { + "name": "model.layers.55.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1703936 + }, + { + "name": "model.layers.55.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1720320 + }, + { + "name": "model.layers.56.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1736704 + }, + { + "name": "model.layers.56.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1753088 + }, + { + "name": "model.layers.57.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1769472 + }, + { + "name": "model.layers.57.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1785856 + }, + { + "name": "model.layers.58.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1802240 + }, + { + "name": "model.layers.58.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1818624 + }, + { + "name": "model.layers.59.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1835008 + }, + { + "name": "model.layers.59.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1851392 + }, + { + "name": "model.layers.60.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1867776 + }, + { + "name": "model.layers.60.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1884160 + }, + { + "name": "model.layers.61.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1900544 + }, + { + "name": "model.layers.61.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1916928 + }, + { + "name": "model.layers.62.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1933312 + }, + { + "name": "model.layers.62.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1949696 + }, + { + "name": "model.layers.63.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1966080 + }, + { + "name": "model.layers.63.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1982464 + }, + { + "name": "model.layers.64.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 1998848 + }, + { + "name": "model.layers.64.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2015232 + }, + { + "name": "model.layers.65.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2031616 + }, + { + "name": "model.layers.65.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2048000 + }, + { + "name": "model.layers.66.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2064384 + }, + { + "name": "model.layers.66.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2080768 + }, + { + "name": "model.layers.67.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2097152 + }, + { + "name": "model.layers.67.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2113536 + }, + { + "name": "model.layers.68.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2129920 + }, + { + "name": "model.layers.68.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2146304 + }, + { + "name": "model.layers.69.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2162688 + }, + { + "name": "model.layers.69.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2179072 + }, + { + "name": "model.layers.70.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2195456 + }, + { + "name": "model.layers.70.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2211840 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2228224 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2244608 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2260992 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2277376 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 8192 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 16384, + "byteOffset": 2293760 + }, + { + "name": "model.layers.9.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": "737217eca43e74b8b038e50486995ce2" + } + ] +} \ No newline at end of file