config.json

{
    "model_type": "llama",
    "hidden_size": 4096,
    "intermediate_size": 11008,
    "num_attention_heads": 32,
    "num_hidden_layers": 32,
    "vocab_size": 32000,
    "max_position_embeddings": 2048,
    "rope_scaling": {
        "type": "dynamic",
        "factor": 2
    },
    "rotary_embedding_base": 10000,
    "tie_word_embeddings": false,
    "torch_dtype": "float16",
    "use_cache": true,
    "pad_token_id": 0,
    "bos_token_id": 1,
    "eos_token_id": 2,
    "quantization_config": {
        "load_in_4bit": true,
        "bnb_4bit_use_double_quant": true,
        "bnb_4bit_quant_type": "nf4"
    }
}