config.json

{
  "architectures": [
    "Transformer"
  ],
  "model_type": "minimind",
  "hidden_size": 512,
  "num_hidden_layers": 8,
  "num_attention_heads": 16,
  "vocab_size": 6400,
  "max_position_embeddings": 512,
  "intermediate_size": 2048,
  "hidden_dropout_prob": 0.0,
  "attention_probs_dropout_prob": 0.0,
  "initializer_range": 0.02,
  "layer_norm_eps": 1e-12,
  "pad_token_id": 0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "use_cache": true
}