config.json

{
  "add_cross_attention": false,
  "architectures": [
    "DPLMForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.0,
  "attn_backend": "sdpa",
  "auto_map": {
    "AutoConfig": "modeling_dplm.DPLMConfig",
    "AutoModel": "modeling_dplm.DPLMModel",
    "AutoModelForMaskedLM": "modeling_dplm.DPLMForMaskedLM",
    "AutoModelForSequenceClassification": "modeling_dplm.DPLMForSequenceClassification",
    "AutoModelForTokenClassification": "modeling_dplm.DPLMForTokenClassification"
  },
  "classifier_dropout": null,
  "dplm_type": "dplm_esm",
  "dtype": "float32",
  "emb_layer_norm_before": false,
  "esmfold_config": null,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 10240,
  "is_decoder": false,
  "is_folding_model": false,
  "layer_norm_eps": 1e-05,
  "mask_token_id": 32,
  "max_position_embeddings": 1026,
  "model_type": "dplm",
  "num_attention_heads": 40,
  "num_hidden_layers": 36,
  "pad_token_id": 1,
  "position_embedding_type": "rotary",
  "tie_word_embeddings": false,
  "token_dropout": true,
  "transformers_version": "5.2.0",
  "use_cache": true,
  "vocab_list": null,
  "vocab_size": 33
}