{ "_name_or_path": "/opt/data/private/EditModel/multi_task/65_10cls_1235_sparse_ffd_id_4head_sent_0.7temper-10epo-5e-5-norm1_enc-dec-nltk616", "architectures": [ "BertForMTSparseFFDEditModel" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "gate_temperature": 0.7, "gate_type": "task_id", "gradient_checkpointing": false, "head_num": 6, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "id2label": { "0": "$APPEND", "1": "$DELETE", "2": "$KEEP", "3": "$MERGE_HYPHEN", "4": "$REPLACE", "5": "$TRANSFORM_AGREEMENT_PLURAL", "6": "$TRANSFORM_AGREEMENT_SINGULAR", "7": "$TRANSFORM_CASE_CAPITAL", "8": "$TRANSFORM_CASE_CAPITAL_1", "9": "$TRANSFORM_CASE_LOWER", "10": "$TRANSFORM_CASE_UPPER", "11": "$TRANSFORM_CASE_UPPER_-1", "12": "$TRANSFORM_SPLIT_HYPHEN", "13": "$TRANSFORM_VERB-" }, "initializer_range": 0.02, "intermediate_size": 3072, "label2id": { "$APPEND": 0, "$DELETE": 1, "$KEEP": 2, "$MERGE_HYPHEN": 3, "$REPLACE": 4, "$TRANSFORM_AGREEMENT_PLURAL": 5, "$TRANSFORM_AGREEMENT_SINGULAR": 6, "$TRANSFORM_CASE_CAPITAL": 7, "$TRANSFORM_CASE_CAPITAL_1": 8, "$TRANSFORM_CASE_LOWER": 9, "$TRANSFORM_CASE_UPPER": 10, "$TRANSFORM_CASE_UPPER_-1": 11, "$TRANSFORM_SPLIT_HYPHEN": 12, "$TRANSFORM_VERB-": 13 }, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "bert", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "sparse_encdec": true, "sparse_level": "sentence_level", "sparse_mode": "ffd", "sum_layers": 12, "sum_tasks": 6, "torch_dtype": "float32", "transformers_version": "4.27.0", "type_vocab_size": 2, "use_cache": true, "vocab_size": 29001 }