config.json

{
    "architectures": [
        "Qwen2ForCausalLM"
    ],
    "attention_dropout": 0.0,
    "bos_token_id": 151646,
    "eos_token_id": 151643,
    "hidden_act": "silu",
    "hidden_size": 3584,
    "initializer_range": 0.02,
    "intermediate_size": 18944,
    "max_position_embeddings": 131072,
    "max_window_layers": 28,
    "model_type": "qwen2",
    "num_attention_heads": 28,
    "num_hidden_layers": 28,
    "num_key_value_heads": 4,
    "pad_token_id": 151643,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 10000,
    "sliding_window": null,
    "tie_word_embeddings": false,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.51.3",
    "use_cache": true,
    "use_mrope": false,
    "use_sliding_window": false,
    "vocab_size": 152064,
    "quantization_config": {
        "ignore": [
            "lm_head"
        ],
        "quant_algo": "NVFP4",
        "kv_cache_scheme": "FP8",
        "producer": {
            "name": "modelopt",
            "version": "0.31.0"
        }
    }
}