{ "_name_or_path": "microsoft/deberta-v3-base", "architectures": [ "DebertaV2ForSequenceClassification" ], "attention_probs_dropout_prob": 0.1, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "id2label": { "0": "pandas", "1": "python-3.x", "2": "string", "3": "django", "4": "dictionary", "5": "python-2.7", "6": "matplotlib", "7": "list", "8": "numpy", "9": "regex" }, "initializer_range": 0.02, "intermediate_size": 3072, "label2id": { "dictionary": 4, "django": 3, "list": 7, "matplotlib": 6, "numpy": 8, "pandas": 0, "python-2.7": 5, "python-3.x": 1, "regex": 9, "string": 2 }, "layer_norm_eps": 1e-07, "max_position_embeddings": 512, "max_relative_positions": -1, "model_type": "deberta-v2", "norm_rel_ebd": "layer_norm", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "pooler_dropout": 0, "pooler_hidden_act": "gelu", "pooler_hidden_size": 768, "pos_att_type": [ "p2c", "c2p" ], "position_biased_input": false, "position_buckets": 256, "problem_type": "multi_label_classification", "relative_attention": true, "share_att_key": true, "torch_dtype": "float32", "transformers_version": "4.45.1", "type_vocab_size": 0, "vocab_size": 128100 }