config.json

{
  "_attn_implementation_autoset": true,
  "architectures": [
    "BertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "auto_map": {
    "AutoConfig": "geneformer.TEBertConfig",
    "AutoModel": "geneformer.BertModel",
    "AutoModelForMaskedLM": "geneformer.BertForMaskedLM"
  },
  "classifier_dropout": null,
  "framework": "pytorch",
  "fuse_qkv_params": true,
  "hidden_act": "relu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 4096,
  "micro_batch_size": null,
  "model_type": "bert",
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "pad_token_id": 0,
  "position_embedding_type": "absolute",
  "torch_dtype": "float32",
  "transformers_version": "4.51.3",
  "type_vocab_size": 2,
  "use_cache": true,
  "use_te_layers": true,
  "vocab_size": 20275
}