config.json

{
  "add_bias_fnn": false,
  "architectures": [
    "EsmForTokenClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "auto_map": {
    "AutoConfig": "esm_config.EsmConfig",
    "AutoModelForMaskedLM": "modeling_esm.EsmForMaskedLM",
    "AutoModelForSequenceClassification": "modeling_esm.EsmForSequenceClassification",
    "AutoModelForTokenClassification": "modeling_esm.EsmForTokenClassification"
  },
  "emb_layer_norm_before": false,
  "esmfold_config": null,
  "hidden_dropout_prob": 0.0,
  "hidden_size": 512,
  "id2label": {
    "0": "O",
    "1": "B-Intron",
    "2": "I-Intron",
    "3": "B-tRNA",
    "4": "I-tRNA",
    "5": "B-anti",
    "6": "I-anti"
  },
  "initializer_range": 0.02,
  "intermediate_size": 2048,
  "is_folding_model": false,
  "label2id": {
    "B-Intron": 1,
    "B-anti": 5,
    "B-tRNA": 3,
    "I-Intron": 2,
    "I-anti": 6,
    "I-tRNA": 4,
    "O": 0
  },
  "layer_norm_eps": 1e-12,
  "mask_token_id": 2,
  "max_position_embeddings": 2050,
  "model_type": "esm",
  "num_attention_heads": 16,
  "num_hidden_layers": 22,
  "pad_token_id": 1,
  "position_embedding_type": "rotary",
  "problem_type": "single_label_classification",
  "tie_word_embeddings": false,
  "token_dropout": false,
  "torch_dtype": "float32",
  "transformers_version": "4.50.3",
  "use_cache": false,
  "vocab_list": null,
  "vocab_size": 11
}