{
    "_name_or_path": "MobileLLM",
    "architectures": [
        "QuantMobileLLM"
    ],
    "model_type": "mobile_llm",
    "vocab_size": 50304,
    "hidden_size": 256,
    "num_attention_heads": 8,
    "num_key_value_heads": 4,
    "num_hidden_layers": 6,
    "intermediate_size": 682,
    "max_position_embeddings": 512,
    "rotary_embeddings_base": 10000.0,
    "tie_word_embeddings": true,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.42.0",
    "use_cache": true
}