| { | |
| "attention_probs_dropout_prob": 0.1, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 3584, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 14336, | |
| "layer_norm_eps": 1e-12, | |
| "max_position_embeddings": 512, | |
| "model_type": "megatron-bert", | |
| "num_attention_heads": 56, | |
| "num_hidden_layers": 56, | |
| "num_layers": 56, | |
| "pad_token_id": 0, | |
| "position_embedding_type": "absolute", | |
| "tokenizer_type": "BertWordPieceLowerCase", | |
| "transformers_version": "4.23.1", | |
| "type_vocab_size": 2, | |
| "use_cache": true, | |
| "vocab_size": 50176 | |
| } | |