{ "metadata": { "ParamSize": 518, "ParamBytes": 1563781120.0, "BitsPerParam": 4.506794107285474 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 64552960, "records": [ { "name": "gpt_neox.embed_in.q_weight", "shape": [ 50432, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 64552960, "byteOffset": 0 } ], "md5sum": "01055e9fb1f7549b971a33d04611c34a" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 22855680, "records": [ { "name": "gpt_neox.embed_in.q_scale", "shape": [ 50432, 80 ], "dtype": "float16", "format": "raw", "nbytes": 8069120, "byteOffset": 0 }, { "name": "gpt_neox.layers.0.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 8069120 }, { "name": "gpt_neox.layers.0.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 8074240 }, { "name": "gpt_neox.layers.0.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 8079360 }, { "name": "gpt_neox.layers.0.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 8084480 }, { "name": "gpt_neox.layers.0.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 8089600 }, { "name": "gpt_neox.layers.0.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 17920000 }, { "name": "gpt_neox.layers.0.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 19148800 }, { "name": "gpt_neox.layers.0.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 19164160 }, { "name": "gpt_neox.layers.0.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 22440960 }, { "name": "gpt_neox.layers.0.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 22850560 } ], "md5sum": "6c73cbcd6b3124836ed04692250ce65e" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.0.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.0.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.0.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.0.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.0.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.0.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.1.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.1.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.1.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.1.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "e9e0bfa74f93027a832bf0c18a0cb947" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.1.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.1.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.1.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.1.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.1.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.1.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.1.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.1.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.1.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "5fe6216672d9c194da2a9f09004c8eaf" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.1.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.1.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.1.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.2.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.2.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.2.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.2.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.2.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.2.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.2.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.2.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.2.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.2.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "4e7de4e40360dbcc99e83b444f6842fe" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.2.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.2.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.2.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.2.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.2.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.2.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.3.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.3.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.3.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.3.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "7239f7899991953e76a30e5cf8ba790d" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.3.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.3.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.3.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.3.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.3.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.3.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.3.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.3.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.3.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "3e77e3c822aae7af4f1efe9290ab42db" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.3.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.3.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.3.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.4.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.4.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.4.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.4.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.4.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.4.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.4.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.4.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.4.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.4.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "b9f3af0cbe31edbcb2f4cca49dec6ddb" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.4.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.4.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.4.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.4.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.4.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.4.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.5.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.5.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.5.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.5.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "9f491cbe69ed7ac3a3e8057aa63033ec" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.5.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.5.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.5.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.5.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.5.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.5.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.5.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.5.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.5.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "9954b35900a54ce0633bb545ef2b1ff0" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.5.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.5.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.5.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.6.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.6.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.6.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.6.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.6.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.6.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.6.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.6.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.6.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.6.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "6518a9b3fd1cceac3628871e85776df4" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.6.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.6.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.6.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.6.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.6.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.6.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.7.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.7.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.7.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.7.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "79e2039ac4d3aa32fa23292693e284eb" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.7.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.7.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.7.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.7.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.7.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.7.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.7.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.7.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.7.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "3631cf04dbed2bbdbde7e8742b2dffc8" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.7.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.7.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.7.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.8.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.8.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.8.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.8.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.8.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.8.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.8.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.8.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.8.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.8.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "0b66f6136923e42705cd6137ae481977" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.8.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.8.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.8.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.8.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.8.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.8.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.9.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.9.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.9.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.9.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "c3cdedc3970b2169e82917e0397ae652" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.9.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.9.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.9.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.9.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.9.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.9.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.9.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.9.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.9.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "bc8f2af9e8bef0611864a6cc0462a185" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.9.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.9.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.9.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.10.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.10.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.10.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.10.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.10.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.10.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.10.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.10.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.10.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.10.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "8f2d072ff4b3bf9ec061f4b9b8228978" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.10.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.10.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.10.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.10.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.10.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.10.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.11.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.11.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.11.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.11.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "dec243e6eb25dfd75e0e2c2caccca9cf" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.11.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.11.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.11.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.11.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.11.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.11.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.11.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.11.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.11.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "18b7021b4eeae5edbc5c6779ea3356d0" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.11.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.11.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.11.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.12.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.12.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.12.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.12.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.12.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.12.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.12.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.12.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.12.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.12.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "9fb37706ced9a884d21d530a118c9258" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.12.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.12.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.12.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.12.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.12.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.12.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.13.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.13.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.13.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.13.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "3a35603271b4926af676a2d2a7012e14" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.13.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.13.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.13.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.13.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.13.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.13.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.13.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.13.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.13.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "741fa5e37a3d52cf2b87848b5e6041b4" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.13.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.13.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.13.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.14.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.14.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.14.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.14.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.14.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.14.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.14.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.14.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.14.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.14.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "8ae89a5f29bb2b9a3ce88b70cbc93408" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.14.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.14.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.14.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.14.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.14.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.14.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.15.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.15.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.15.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.15.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "0137680d61705e63dca4b404903f4563" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.15.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.15.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.15.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.15.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.15.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.15.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.15.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.15.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.15.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "934d66de5deb53309ea644fa443379a2" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.15.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.15.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.15.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.16.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.16.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.16.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.16.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.16.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.16.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.16.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.16.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.16.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.16.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "9e1f553f564276a9df4dc41f004cd47b" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.16.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.16.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.16.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.16.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.16.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.16.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.17.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.17.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.17.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.17.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "8dbd85e045b65e00c31e95f2e1a94fd3" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.17.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.17.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.17.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.17.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.17.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.17.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.17.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.17.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.17.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "43bc45fae5c7d736085db7b81c56ed56" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.17.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.17.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.17.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.18.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.18.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.18.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.18.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.18.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.18.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.18.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.18.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.18.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.18.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "d0764b3631c19056a61038477b45106d" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.18.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.18.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.18.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.18.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.18.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.18.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.19.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.19.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.19.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.19.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "f459591396cf72fa0426984cbd1f4a27" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.19.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.19.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.19.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.19.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.19.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.19.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.19.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.19.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.19.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "ebb4bed985bee95be0faff0adab27e33" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.19.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.19.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.19.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.20.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.20.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.20.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.20.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.20.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.20.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.20.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.20.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.20.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.20.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "eb0637dac00aa88b28ca173fa6ba4514" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.20.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.20.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.20.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.20.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.20.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.20.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.21.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.21.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.21.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.21.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "3733a4cfa71822c0110baf68a595926f" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.21.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.21.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.21.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.21.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.21.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.21.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.21.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.21.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.21.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "fabb079e335a76761c3811c8e59fdee7" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.21.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.21.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.21.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.22.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.22.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.22.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.22.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.22.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.22.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.22.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.22.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.22.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.22.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "1b521cdb09f61b58d27e95e4e811c1de" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.22.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.22.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.22.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.22.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.22.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.22.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.23.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.23.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.23.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.23.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "042d5f9f98923415f64c558b87b68941" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.23.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.23.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.23.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.23.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.23.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.23.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.23.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.23.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.23.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "b33827050c122dcae634d9f1acd6d3b1" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.23.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.23.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.23.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.24.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.24.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.24.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.24.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.24.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.24.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.24.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.24.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.24.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.24.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "5c539c3eb8c2c4aab5097174dd65f731" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.24.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.24.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.24.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.24.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.24.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.24.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.25.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.25.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.25.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.25.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "c8f5a9aff1cc5188d1ee311034e2ac5c" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.25.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.25.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.25.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.25.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.25.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.25.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.25.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.25.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.25.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "5eed10c610a6a3dd8deef2b54f4bea40" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.25.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.25.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.25.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.26.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.26.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.26.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.26.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.26.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.26.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.26.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.26.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.26.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.26.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "ea649eb70ebe7741a54af59e8320ec39" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.26.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.26.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.26.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.26.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.26.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.26.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.27.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.27.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.27.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.27.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "a148600cf71d18b2120c11150fbb2ef1" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.27.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.27.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.27.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.27.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.27.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.27.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.27.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.27.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.27.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "342141caf9ed6563f2894ee34dc1458c" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.27.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.27.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.27.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.28.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.28.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.28.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.28.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.28.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.28.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.28.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.28.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.28.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.28.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "c739228f464797731934f711a08bd1fd" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.28.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.28.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.28.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.28.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.28.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.28.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.29.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.29.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.29.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.29.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "894d92aea73e12f1f4f2c23aaab1daa7" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.29.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.29.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.29.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.29.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.29.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.29.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.29.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.29.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.29.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "fe5ab0791f18fdeb18e8d83ec5156900" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 29542400, "records": [ { "name": "gpt_neox.layers.29.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.29.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.29.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.30.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.layers.30.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.30.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.30.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14771200 }, { "name": "gpt_neox.layers.30.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 14776320 }, { "name": "gpt_neox.layers.30.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 24606720 }, { "name": "gpt_neox.layers.30.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 25835520 }, { "name": "gpt_neox.layers.30.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 25850880 }, { "name": "gpt_neox.layers.30.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 29127680 }, { "name": "gpt_neox.layers.30.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29537280 } ], "md5sum": "3385628ca0f06762650b1b01317981aa" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 29562880, "records": [ { "name": "gpt_neox.layers.30.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.30.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.30.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 14745600 }, { "name": "gpt_neox.layers.30.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14786560 }, { "name": "gpt_neox.layers.30.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27893760 }, { "name": "gpt_neox.layers.30.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 29532160 }, { "name": "gpt_neox.layers.31.input_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29542400 }, { "name": "gpt_neox.layers.31.input_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29547520 }, { "name": "gpt_neox.layers.31.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29552640 }, { "name": "gpt_neox.layers.31.post_attention_layernorm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 29557760 } ], "md5sum": "193d4038a5bcfec951942b4d834acd56" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 29552640, "records": [ { "name": "gpt_neox.layers.31.attention.query_key_value.q_weight", "shape": [ 7680, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 9830400, "byteOffset": 0 }, { "name": "gpt_neox.layers.31.attention.query_key_value.q_scale", "shape": [ 7680, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1228800, "byteOffset": 9830400 }, { "name": "gpt_neox.layers.31.attention.query_key_value.bias", "shape": [ 7680 ], "dtype": "float16", "format": "raw", "nbytes": 15360, "byteOffset": 11059200 }, { "name": "gpt_neox.layers.31.attention.dense.q_weight", "shape": [ 2560, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 3276800, "byteOffset": 11074560 }, { "name": "gpt_neox.layers.31.attention.dense.q_scale", "shape": [ 2560, 80 ], "dtype": "float16", "format": "raw", "nbytes": 409600, "byteOffset": 14351360 }, { "name": "gpt_neox.layers.31.attention.dense.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "gpt_neox.layers.31.mlp.dense_h_to_4h.q_weight", "shape": [ 10240, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "gpt_neox.layers.31.mlp.dense_h_to_4h.q_scale", "shape": [ 10240, 80 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "gpt_neox.layers.31.mlp.dense_h_to_4h.bias", "shape": [ 10240 ], "dtype": "float32", "format": "raw", "nbytes": 40960, "byteOffset": 29511680 } ], "md5sum": "860f45b0569c3955d160bbe56b359c8b" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 64552960, "records": [ { "name": "embed_out.q_weight", "shape": [ 50432, 320 ], "dtype": "uint32", "format": "raw", "nbytes": 64552960, "byteOffset": 0 } ], "md5sum": "3e6e9aa679bf3ac1083721e97c3cb0f6" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 22835200, "records": [ { "name": "gpt_neox.layers.31.mlp.dense_4h_to_h.q_weight", "shape": [ 2560, 1280 ], "dtype": "uint32", "format": "raw", "nbytes": 13107200, "byteOffset": 0 }, { "name": "gpt_neox.layers.31.mlp.dense_4h_to_h.q_scale", "shape": [ 2560, 320 ], "dtype": "float16", "format": "raw", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "gpt_neox.layers.31.mlp.dense_4h_to_h.bias", "shape": [ 2560 ], "dtype": "float32", "format": "raw", "nbytes": 10240, "byteOffset": 14745600 }, { "name": "gpt_neox.final_layer_norm.weight", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "gpt_neox.final_layer_norm.bias", "shape": [ 2560 ], "dtype": "float16", "format": "raw", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "embed_out.q_scale", "shape": [ 50432, 80 ], "dtype": "float16", "format": "raw", "nbytes": 8069120, "byteOffset": 14766080 } ], "md5sum": "8e11f9890e002728ea73cb7b893684bf" } ] }