{ "_name_or_path": "/opt/ml/ray_results/_objective_2023-01-03_10-02-38/_objective_c119e_00000_0_learning_rate=0.0000,num_train_epochs=5,per_device_train_batch_size=16_2023-01-03_10-02-38/saved_models/klue/roberta-large/mrc_LSG_01-03-19-01/run-c119e_00000/checkpoint-21785", "architectures": [ "RobertaForQuestionAnswering" ], "attention_probs_dropout_prob": 0.1, "bos_token_id": 0, "classifier_dropout": null, "eos_token_id": 2, "gradient_checkpointing": false, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 1024, "initializer_range": 0.02, "intermediate_size": 4096, "layer_norm_eps": 1e-05, "max_position_embeddings": 514, "model_type": "roberta", "num_attention_heads": 16, "num_hidden_layers": 24, "pad_token_id": 1, "position_embedding_type": "absolute", "tokenizer_class": "BertTokenizer", "torch_dtype": "float32", "transformers_version": "4.25.1", "type_vocab_size": 1, "use_cache": true, "vocab_size": 32000 }