| { | |
| "architectures": ["RobertaForSequenceClassification"], | |
| "model_type": "roberta", | |
| "num_classes": 2, | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 24, | |
| "hidden_size": 1024, | |
| "intermediate_size": 4096, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "attention_probs_dropout_prob": 0.1, | |
| "max_position_embeddings": 514, | |
| "type_vocab_size": 1, | |
| "initializer_range": 0.02, | |
| "layer_norm_eps": 1e-05, | |
| "classifier_dropout": null, | |
| "task_specific_params": { | |
| "sentence_prediction": { | |
| "dropout": 0.1, | |
| "classifier_dropout": 0.1 | |
| } | |
| }, | |
| "lr": 1e-05, | |
| "total_num_updates": 7812, | |
| "warmup_updates": 469, | |
| "classification_head_name": "imdb_head", | |
| "optimizer": "adam", | |
| "adam_betas": [0.9, 0.98], | |
| "adam_eps": 1e-06, | |
| "weight_decay": 0.1, | |
| "learning_rate_scheduler": "polynomial_decay", | |
| "max_sentences": 32, | |
| "max_tokens": 4400, | |
| "batch_size": 32, | |
| "init_token": 0, | |
| "separator_token": 2 | |
| } | |