config.json

{
  "architectures": [
    "CustomTransformerModel"
  ],
  "bos_token_id": 2,
  "d_ff": 3072,
  "d_model": 768,
  "dropout": 0.1,
  "eos_token_id": 3,
  "max_position_embeddings": 512,
  "model_type": "miscovery",
  "num_decoder_layers": 12,
  "num_encoder_layers": 12,
  "num_heads": 12,
  "pad_token_id": 0,
  "torch_dtype": "float32",
  "transformers_version": "4.51.3",
  "use_flash_attn": true,
  "vocab_size": 100000
}