{ "_name_or_path": "output_dir_XTBert", "alibi_starting_size": 512, "architectures": [ "BertForSequenceClassification" ], "attention_probs_dropout_prob": 0.0, "auto_map": { "AutoConfig": "configuration_bert.BertConfig", "AutoModelForMaskedLM": "xiaotinghe/XTBert--bert_layers.BertForMaskedLM", "AutoModelForSequenceClassification": "bert_layers.BertForSequenceClassification" }, "bos_token_id": 0, "classifier_dropout": null, "directionality": "bidi", "eos_token_id": 2, "gradient_checkpointing": false, "hidden_act": "silu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "id2label": { "0": "academic disciplines", "1": "business", "2": "code", "3": "communication", "4": "culture", "5": "economy", "6": "education", "7": "energy", "8": "engineering", "9": "entertainment", "10": "food and drink", "11": "geography", "12": "government", "13": "history", "14": "human behavior", "15": "humanities", "16": "information", "17": "internet", "18": "knowledge", "19": "language", "20": "law", "21": "life health", "22": "mass media", "23": "mathematics", "24": "military", "25": "nature", "26": "people", "27": "philosophy", "28": "politics", "29": "religion", "30": "science", "31": "society", "32": "sports", "33": "time" }, "initializer_range": 0.02, "intermediate_size": 2048, "label2id": { "academic disciplines": 0, "business": 1, "code": 2, "communication": 3, "culture": 4, "economy": 5, "education": 6, "energy": 7, "engineering": 8, "entertainment": 9, "food and drink": 10, "geography": 11, "government": 12, "history": 13, "human behavior": 14, "humanities": 15, "information": 16, "internet": 17, "knowledge": 18, "language": 19, "law": 20, "life health": 21, "mass media": 22, "mathematics": 23, "military": 24, "nature": 25, "people": 26, "philosophy": 27, "politics": 28, "religion": 29, "science": 30, "society": 31, "sports": 32, "time": 33 }, "layer_norm_eps": 1e-12, "max_position_embeddings": 4096, "model_type": "bert", "num_attention_heads": 12, "num_hidden_layers": 12, "output_past": true, "pad_token_id": 1, "pooler_fc_size": 768, "pooler_num_attention_heads": 12, "pooler_num_fc_layers": 3, "pooler_size_per_head": 128, "pooler_type": "first_token_transform", "position_embedding_type": "absolute", "problem_type": "single_label_classification", "torch_dtype": "float32", "transformers_version": "4.33.2", "type_vocab_size": 2, "use_cache": true, "vocab_size": 39984 }