{
  "architectures": [
    "NeoBERTLMHead"
  ],
  "decoder_init_range": 0.02,
  "dim_head": 88,
  "embedding_init_range": 0.02,
  "hidden_size": 704,
  "intermediate_size": 2112,
  "kwargs": {},
  "max_length": 256,
  "model_type": "neobert",
  "norm_eps": 1e-06,
  "num_attention_heads": 8,
  "num_hidden_layers": 8,
  "pad_token_id": 0,
  "torch_dtype": "float32",
  "transformers_version": "4.45.2",
  "vocab_size": 30522
}