{
  "architectures": [
    "TransformerLMForCausalLM"
  ],
  "attention_dropout_rate": 0.1,
  "auto_map": {
    "AutoConfig": "configuration_transformerlm.TransformerLMConfig",
    "FlaxAutoModelForCausalLM": "modeling_transformerlm_flax.FlaxTransformerLMForCausalLM"
  },
  "bos_token_id": 50256,
  "decode": true,
  "deterministic": true,
  "dropout_rate": 0.1,
  "emb_dim": 2048,
  "eos_token_id": 50256,
  "logits_via_embedding": false,
  "max_len": 1024,
  "mlp_dim": 8192,
  "model_type": "transformerlm",
  "num_heads": 16,
  "num_layers": 18,
  "output_vocab_size": 30000,
  "qkv_dim": 2048,
  "share_embeddings": false,
  "tokenizer_class": "TransformerLMTokenizer",
  "transformers_version": "4.39.0",
  "vocab_size": 30000
}