config.json

{
  "architectures": [
    "ProkBertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "auto_map": {
    "AutoConfig": "neuralbioinfo/nbrg-transformers--models.ProkBertConfig",
    "AutoModel": "neuralbioinfo/nbrg-transformers--models.ProkBertModel",
    "AutoModelForMaskedLM": "neuralbioinfo/nbrg-transformers--models.ProkBertForMaskedLM",
    "AutoModelForSequenceClassification": "neuralbioinfo/nbrg-transformers--models.ProkBertForSequenceClassification"
  },
  "classifier_dropout": 0.1,
  "classifier_head_type": "linear",
  "classifier_mlp_hidden_size": null,
  "classifier_pooling": "attention",
  "curricular_embedding_size": null,
  "curricular_margin": 0.5,
  "curricular_scale": 64.0,
  "dtype": "float32",
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 384,
  "initializer_range": 0.02,
  "intermediate_size": 4096,
  "kmer": 6,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 2048,
  "model_type": "prokbert",
  "num_attention_heads": 6,
  "num_hidden_layers": 6,
  "pad_token_id": 0,
  "position_embedding_type": "relative_key_query",
  "shift": 2,
  "transformers_version": "4.57.6",
  "type_vocab_size": 2,
  "use_cache": true,
  "vocab_size": 4200
}