config.json
| 1 | { |
| 2 | "_num_labels": 3, |
| 3 | "activation_dropout": 0.0, |
| 4 | "activation_function": "gelu", |
| 5 | "add_final_layer_norm": false, |
| 6 | "architectures": [ |
| 7 | "BartForSequenceClassification" |
| 8 | ], |
| 9 | "attention_dropout": 0.0, |
| 10 | "bos_token_id": 0, |
| 11 | "classif_dropout": 0.0, |
| 12 | "classifier_dropout": 0.0, |
| 13 | "d_model": 1024, |
| 14 | "decoder_attention_heads": 16, |
| 15 | "decoder_ffn_dim": 4096, |
| 16 | "decoder_layerdrop": 0.0, |
| 17 | "decoder_layers": 12, |
| 18 | "decoder_start_token_id": 2, |
| 19 | "dropout": 0.1, |
| 20 | "encoder_attention_heads": 16, |
| 21 | "encoder_ffn_dim": 4096, |
| 22 | "encoder_layerdrop": 0.0, |
| 23 | "encoder_layers": 12, |
| 24 | "eos_token_id": 2, |
| 25 | "forced_eos_token_id": 2, |
| 26 | "gradient_checkpointing": false, |
| 27 | "id2label": { |
| 28 | "0": "contradiction", |
| 29 | "1": "neutral", |
| 30 | "2": "entailment" |
| 31 | }, |
| 32 | "init_std": 0.02, |
| 33 | "is_encoder_decoder": true, |
| 34 | "label2id": { |
| 35 | "contradiction": 0, |
| 36 | "entailment": 2, |
| 37 | "neutral": 1 |
| 38 | }, |
| 39 | "max_position_embeddings": 1024, |
| 40 | "model_type": "bart", |
| 41 | "normalize_before": false, |
| 42 | "num_hidden_layers": 12, |
| 43 | "output_past": false, |
| 44 | "pad_token_id": 1, |
| 45 | "scale_embedding": false, |
| 46 | "transformers_version": "4.7.0.dev0", |
| 47 | "use_cache": true, |
| 48 | "vocab_size": 50265 |
| 49 | } |
| 50 | |