config.json

{
    "producer": {
        "name": "modelopt",
        "version": "0.15.1"
    },
    "architecture": "LlamaForCausalLM",
    "dtype": "bfloat16",
    "logits_dtype": "float16",
    "num_hidden_layers": 32,
    "num_attention_heads": 32,
    "num_key_value_heads": 8,
    "hidden_size": 4096,
    "norm_epsilon": 1e-05,
    "vocab_size": 128256,
    "max_position_embeddings": 131072,
    "hidden_act": "silu",
    "use_parallel_embedding": true,
    "embedding_sharding_dim": 0,
    "quantization": {
        "quant_algo": "W4A16_AWQ",
        "kv_cache_quant_algo": null,
        "group_size": 64,
        "has_zero_point": false,
        "pre_quant_scale": true,
        "exclude_modules": [
            "lm_head"
        ]
    },
    "mapping": {
        "world_size": 1,
        "tp_size": 1,
        "pp_size": 1
    },
    "head_size": 128,
    "intermediate_size": 14336,
    "position_embedding_type": "rope_gpt_neox",
    "share_embedding_table": false,
    "residual_mlp": false,
    "bias": false,
    "rotary_pct": 1.0,
    "rank": 0,
    "decoder": "llama",
    "rmsnorm": true,
    "lm_head_bias": false,
    "rotary_base": 500000.0,
    "rotary_scaling": {
        "factor": 8.0,
        "low_freq_factor": 1.0,
        "high_freq_factor": 4.0,
        "original_max_position_embeddings": 8192,
        "rope_type": "llama3"
    }
}