| { | |
| "architectures": [ | |
| "ModernBertForSequenceClassification" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 50281, | |
| "classifier_activation": "gelu", | |
| "classifier_bias": false, | |
| "classifier_dropout": 0.0, | |
| "classifier_pooling": "mean", | |
| "cls_token_id": 50281, | |
| "decoder_bias": true, | |
| "deterministic_flash_attn": false, | |
| "embedding_dropout": 0.0, | |
| "eos_token_id": 50282, | |
| "global_attn_every_n_layers": 3, | |
| "global_rope_theta": 160000.0, | |
| "gradient_checkpointing": false, | |
| "hidden_activation": "gelu", | |
| "hidden_size": 1024, | |
| "id2label": { | |
| "0": "NA", | |
| "1": "Incomplete", | |
| "2": "WNB", | |
| "3": "SwapDividend", | |
| "4": "Mult", | |
| "5": "FlipChange", | |
| "6": "Irrelevant", | |
| "7": "Wrong_Fraction", | |
| "8": "Additive", | |
| "9": "Not_variable", | |
| "10": "Adding_terms", | |
| "11": "Inverse_operation", | |
| "12": "Inversion", | |
| "13": "Duplication", | |
| "14": "Wrong_Operation", | |
| "15": "Whole_numbers_larger", | |
| "16": "Longer_is_bigger", | |
| "17": "Ignores_zeroes", | |
| "18": "Shorter_is_bigger", | |
| "19": "Wrong_fraction", | |
| "20": "Adding_across", | |
| "21": "Denominator-only_change", | |
| "22": "Incorrect_equivalent_fraction_addition", | |
| "23": "Division", | |
| "24": "Subtraction", | |
| "25": "Unknowable", | |
| "26": "Definition", | |
| "27": "Interior", | |
| "28": "Positive", | |
| "29": "Tacking", | |
| "30": "Wrong_term", | |
| "31": "Firstterm", | |
| "32": "Base_rate", | |
| "33": "Multiplying_by_4", | |
| "34": "Certainty", | |
| "35": "Scale" | |
| }, | |
| "initializer_cutoff_factor": 2.0, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 2624, | |
| "label2id": { | |
| "Adding_across": 20, | |
| "Adding_terms": 10, | |
| "Additive": 8, | |
| "Base_rate": 32, | |
| "Certainty": 34, | |
| "Definition": 26, | |
| "Denominator-only_change": 21, | |
| "Division": 23, | |
| "Duplication": 13, | |
| "Firstterm": 31, | |
| "FlipChange": 5, | |
| "Ignores_zeroes": 17, | |
| "Incomplete": 1, | |
| "Incorrect_equivalent_fraction_addition": 22, | |
| "Interior": 27, | |
| "Inverse_operation": 11, | |
| "Inversion": 12, | |
| "Irrelevant": 6, | |
| "Longer_is_bigger": 16, | |
| "Mult": 4, | |
| "Multiplying_by_4": 33, | |
| "NA": 0, | |
| "Not_variable": 9, | |
| "Positive": 28, | |
| "Scale": 35, | |
| "Shorter_is_bigger": 18, | |
| "Subtraction": 24, | |
| "SwapDividend": 3, | |
| "Tacking": 29, | |
| "Unknowable": 25, | |
| "WNB": 2, | |
| "Whole_numbers_larger": 15, | |
| "Wrong_Fraction": 7, | |
| "Wrong_Operation": 14, | |
| "Wrong_fraction": 19, | |
| "Wrong_term": 30 | |
| }, | |
| "layer_norm_eps": 1e-05, | |
| "local_attention": 128, | |
| "local_rope_theta": 10000.0, | |
| "max_position_embeddings": 8192, | |
| "mlp_bias": false, | |
| "mlp_dropout": 0.0, | |
| "model_type": "modernbert", | |
| "norm_bias": false, | |
| "norm_eps": 1e-05, | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 28, | |
| "pad_token_id": 50283, | |
| "position_embedding_type": "absolute", | |
| "problem_type": "single_label_classification", | |
| "repad_logits_with_grad": false, | |
| "sep_token_id": 50282, | |
| "sparse_pred_ignore_index": -100, | |
| "sparse_prediction": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.55.4", | |
| "unsloth_version": "2025.10.1", | |
| "vocab_size": 50368 | |
| } | |