{ "metadata": { "ParamSize": 76, "ParamBytes": 281706496.0, "BitsPerParam": 32.0 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 103022592, "records": [ { "name": "embed_out.weight", "shape": [ 50304, 512 ], "dtype": "float32", "format": "raw", "nbytes": 103022592, "byteOffset": 0 } ], "md5sum": "0a623ea8144d86c80bdf32b3f20f378c" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 103022592, "records": [ { "name": "gpt_neox.embed_in.weight", "shape": [ 50304, 512 ], "dtype": "float32", "format": "raw", "nbytes": 103022592, "byteOffset": 0 } ], "md5sum": "f945d3830b66ed726b53093acf4b7510" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 29431808, "records": [ { "name": "gpt_neox.final_layer_norm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 0 }, { "name": "gpt_neox.final_layer_norm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 2048 }, { "name": "gpt_neox.layers.0.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4096 }, { "name": "gpt_neox.layers.0.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 6144 }, { "name": "gpt_neox.layers.0.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 1054720 }, { "name": "gpt_neox.layers.0.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 1060864 }, { "name": "gpt_neox.layers.0.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4206592 }, { "name": "gpt_neox.layers.0.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4208640 }, { "name": "gpt_neox.layers.0.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4210688 }, { "name": "gpt_neox.layers.0.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 4212736 }, { "name": "gpt_neox.layers.0.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 8407040 }, { "name": "gpt_neox.layers.0.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 8415232 }, { "name": "gpt_neox.layers.0.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12609536 }, { "name": "gpt_neox.layers.0.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12611584 }, { "name": "gpt_neox.layers.1.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12613632 }, { "name": "gpt_neox.layers.1.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 12615680 }, { "name": "gpt_neox.layers.1.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 13664256 }, { "name": "gpt_neox.layers.1.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 13670400 }, { "name": "gpt_neox.layers.1.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 16816128 }, { "name": "gpt_neox.layers.1.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 16818176 }, { "name": "gpt_neox.layers.1.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 16820224 }, { "name": "gpt_neox.layers.1.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 16822272 }, { "name": "gpt_neox.layers.1.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 21016576 }, { "name": "gpt_neox.layers.1.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 21024768 }, { "name": "gpt_neox.layers.1.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25219072 }, { "name": "gpt_neox.layers.1.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25221120 }, { "name": "gpt_neox.layers.2.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25223168 }, { "name": "gpt_neox.layers.2.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 25225216 }, { "name": "gpt_neox.layers.2.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 26273792 }, { "name": "gpt_neox.layers.2.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 26279936 }, { "name": "gpt_neox.layers.2.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 29425664 }, { "name": "gpt_neox.layers.2.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 29427712 }, { "name": "gpt_neox.layers.2.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 29429760 } ], "md5sum": "a5b014faa17e4f1a8e65a813eb0dfb3a" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 29421568, "records": [ { "name": "gpt_neox.layers.2.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 0 }, { "name": "gpt_neox.layers.2.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 4194304 }, { "name": "gpt_neox.layers.2.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 4202496 }, { "name": "gpt_neox.layers.2.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8396800 }, { "name": "gpt_neox.layers.2.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8398848 }, { "name": "gpt_neox.layers.3.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8400896 }, { "name": "gpt_neox.layers.3.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 8402944 }, { "name": "gpt_neox.layers.3.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 9451520 }, { "name": "gpt_neox.layers.3.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 9457664 }, { "name": "gpt_neox.layers.3.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12603392 }, { "name": "gpt_neox.layers.3.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12605440 }, { "name": "gpt_neox.layers.3.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 12607488 }, { "name": "gpt_neox.layers.3.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 12609536 }, { "name": "gpt_neox.layers.3.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 16803840 }, { "name": "gpt_neox.layers.3.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 16812032 }, { "name": "gpt_neox.layers.3.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 21006336 }, { "name": "gpt_neox.layers.3.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 21008384 }, { "name": "gpt_neox.layers.4.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 21010432 }, { "name": "gpt_neox.layers.4.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 21012480 }, { "name": "gpt_neox.layers.4.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 22061056 }, { "name": "gpt_neox.layers.4.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 22067200 }, { "name": "gpt_neox.layers.4.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25212928 }, { "name": "gpt_neox.layers.4.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25214976 }, { "name": "gpt_neox.layers.4.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 25217024 }, { "name": "gpt_neox.layers.4.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 25219072 }, { "name": "gpt_neox.layers.4.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 29413376 } ], "md5sum": "8938b1b2f84660600e662265cf728153" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 16807936, "records": [ { "name": "gpt_neox.layers.4.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 0 }, { "name": "gpt_neox.layers.4.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4194304 }, { "name": "gpt_neox.layers.4.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4196352 }, { "name": "gpt_neox.layers.5.attention.dense.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 4198400 }, { "name": "gpt_neox.layers.5.attention.dense.weight", "shape": [ 512, 512 ], "dtype": "float32", "format": "raw", "nbytes": 1048576, "byteOffset": 4200448 }, { "name": "gpt_neox.layers.5.attention.query_key_value.bias", "shape": [ 1536 ], "dtype": "float32", "format": "raw", "nbytes": 6144, "byteOffset": 5249024 }, { "name": "gpt_neox.layers.5.attention.query_key_value.weight", "shape": [ 1536, 512 ], "dtype": "float32", "format": "raw", "nbytes": 3145728, "byteOffset": 5255168 }, { "name": "gpt_neox.layers.5.input_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8400896 }, { "name": "gpt_neox.layers.5.input_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8402944 }, { "name": "gpt_neox.layers.5.mlp.dense_4h_to_h.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 8404992 }, { "name": "gpt_neox.layers.5.mlp.dense_4h_to_h.weight", "shape": [ 512, 2048 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 8407040 }, { "name": "gpt_neox.layers.5.mlp.dense_h_to_4h.bias", "shape": [ 2048 ], "dtype": "float32", "format": "raw", "nbytes": 8192, "byteOffset": 12601344 }, { "name": "gpt_neox.layers.5.mlp.dense_h_to_4h.weight", "shape": [ 2048, 512 ], "dtype": "float32", "format": "raw", "nbytes": 4194304, "byteOffset": 12609536 }, { "name": "gpt_neox.layers.5.post_attention_layernorm.bias", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 16803840 }, { "name": "gpt_neox.layers.5.post_attention_layernorm.weight", "shape": [ 512 ], "dtype": "float32", "format": "raw", "nbytes": 2048, "byteOffset": 16805888 } ], "md5sum": "1647274ac091bba533b982ed2c89a91b" } ] }