CodeIsAbstract/ReasoningTextClassifier
06
1{2 "_name_or_path": "./model",3 "architectures": [4 "BertForSequenceClassification"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "attention_probs_dropout_prob": 0.1,9 "bos_token_id": 50281,10 "classifier_activation": "gelu",11 "classifier_bias": false,12 "classifier_dropout": 0.0,13 "classifier_pooling": "mean",14 "cls_token_id": 50281,15 "decoder_bias": true,16 "deterministic_flash_attn": false,17 "embedding_dropout": 0.0,18 "eos_token_id": 50282,19 "global_attn_every_n_layers": 3,20 "global_rope_theta": 160000.0,21 "gradient_checkpointing": false,22 "hidden_act": "gelu",23 "hidden_activation": "gelu",24 "hidden_dropout_prob": 0.1,25 "hidden_size": 768,26 "id2label": {27 "0": "non-reasoning",28 "1": "reasoning"29 },30 "initializer_cutoff_factor": 2.0,31 "initializer_range": 0.02,32 "intermediate_size": 1152,33 "label2id": {34 "non-reasoning": 0,35 "reasoning": 136 },37 "layer_norm_eps": 1e-05,38 "local_attention": 128,39 "local_rope_theta": 10000.0,40 "max_position_embeddings": 8192,41 "mlp_bias": false,42 "mlp_dropout": 0.0,43 "model_type": "bert",44 "norm_bias": false,45 "norm_eps": 1e-05,46 "num_attention_heads": 12,47 "num_hidden_layers": 22,48 "pad_token_id": 50283,49 "position_embedding_type": "absolute",50 "problem_type": "single_label_classification",51 "sep_token_id": 50282,52 "torch_dtype": "float32",53 "transformers_version": "4.48.3",54 "type_vocab_size": 2,55 "use_cache": true,56 "vocab_size": 5036857}58 