| { | |
| "_name_or_path": "./url_relevance/checkpoint-1000", | |
| "activation": "gelu", | |
| "architectures": [ | |
| "DistilBertForSequenceClassification" | |
| ], | |
| "attention_dropout": 0.1, | |
| "dim": 768, | |
| "dropout": 0.1, | |
| "hidden_dim": 3072, | |
| "initializer_range": 0.02, | |
| "max_position_embeddings": 2048, | |
| "model_type": "distilbert", | |
| "n_heads": 12, | |
| "n_layers": 6, | |
| "pad_token_id": 0, | |
| "qa_dropout": 0.1, | |
| "seq_classif_dropout": 0.2, | |
| "sinusoidal_pos_embds": false, | |
| "tabular_config": { | |
| "cat_feat_dim": 0, | |
| "categorical_bn": true, | |
| "combine_feat_method": "text_only", | |
| "gating_beta": 0.2, | |
| "hidden_dropout_prob": 0.2, | |
| "mlp_act": "relu", | |
| "mlp_division": 4, | |
| "mlp_dropout": 0.1, | |
| "num_labels": 2, | |
| "numerical_bn": true, | |
| "numerical_feat_dim": 0, | |
| "text_feat_dim": 768, | |
| "use_simple_classifier": true | |
| }, | |
| "tie_weights_": true, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.38.2", | |
| "vocab_size": 30522 | |
| } | |