config.json

{
  "architectures": [
    "NQLMModel"
  ],
  "model_type": "nqlm",
  "hidden_size": 4096,
  "intermediate_size": 11008,
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "vocab_size": 32000,
  "max_position_embeddings": 4096,
  "initializer_range": 0.02,
  "layer_norm_eps": 1e-5,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 0,
  "torch_dtype": "bfloat16",
  "use_cache": true,
  "transformers_version": "4.39.0"
}