{ "_name_or_path": "answerdotai/ModernBERT-base", "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": 768, "id2label": { "0": "428", "1": "429", "2": "430", "3": "431", "4": "432", "5": "433", "6": "434", "7": "435", "8": "436", "9": "437", "10": "438", "11": "439", "12": "440", "13": "442", "14": "443", "15": "444", "16": "445", "17": "446", "18": "447", "19": "448", "20": "449", "21": "450", "22": "451", "23": "452", "24": "453", "25": "454", "26": "455", "27": "456", "28": "458", "29": "459", "30": "503" }, "initializer_cutoff_factor": 2.0, "initializer_range": 0.02, "intermediate_size": 1152, "label2id": { "428": 0, "429": 1, "430": 2, "431": 3, "432": 4, "433": 5, "434": 6, "435": 7, "436": 8, "437": 9, "438": 10, "439": 11, "440": 12, "442": 13, "443": 14, "444": 15, "445": 16, "446": 17, "447": 18, "448": 19, "449": 20, "450": 21, "451": 22, "452": 23, "453": 24, "454": 25, "455": 26, "456": 27, "458": 28, "459": 29, "503": 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": 12, "num_hidden_layers": 22, "pad_token_id": 50283, "position_embedding_type": "absolute", "problem_type": "single_label_classification", "reference_compile": true, "sep_token_id": 50282, "sparse_pred_ignore_index": -100, "sparse_prediction": false, "torch_dtype": "float32", "transformers_version": "4.48.0.dev0", "vocab_size": 50368 }