| { | |
| "_name_or_path": "sberbank-ai/sbert_large_nlu_ru", | |
| "architectures": [ | |
| "BertForTokenClassification" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "classifier_dropout": null, | |
| "directionality": "bidi", | |
| "gradient_checkpointing": false, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 1024, | |
| "id2label": { | |
| "1": "LOWER_O", | |
| "2": "LOWER_PERIOD", | |
| "3": "LOWER_COMMA", | |
| "4": "LOWER_QUESTION", | |
| "5": "UPPER_O", | |
| "6": "UPPER_PERIOD", | |
| "7": "UPPER_COMMA", | |
| "8": "UPPER_QUESTION", | |
| "9": "UPPER_TOTAL_O", | |
| "10": "UPPER_TOTAL_PERIOD", | |
| "11": "UPPER_TOTAL_COMMA", | |
| "12": "UPPER_TOTAL_QUESTION", | |
| "-100": "O" | |
| }, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "label2id": { | |
| "LOWER_O": 1, | |
| "LOWER_PERIOD": 2, | |
| "LOWER_COMMA": 3, | |
| "LOWER_QUESTION": 4, | |
| "UPPER_O": 5, | |
| "UPPER_PERIOD": 6, | |
| "UPPER_COMMA": 7, | |
| "UPPER_QUESTION": 8, | |
| "UPPER_TOTAL_O": 9, | |
| "UPPER_TOTAL_PERIOD": 10, | |
| "UPPER_TOTAL_COMMA": 11, | |
| "UPPER_TOTAL_QUESTION": 12, | |
| "O": -100 | |
| }, | |
| "layer_norm_eps": 1e-12, | |
| "max_position_embeddings": 512, | |
| "model_type": "bert", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 24, | |
| "pad_token_id": 0, | |
| "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", | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.36.2", | |
| "type_vocab_size": 2, | |
| "use_cache": true, | |
| "vocab_size": 120138 | |
| } |