{
    "_attn_implementation_autoset": false,
    "architectures": [
        "DeepseekV3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "auto_map": {
        "AutoConfig": "configuration_deepseek.DeepseekV3Config",
        "AutoModel": "modeling_deepseek.DeepseekV3Model",
        "AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
    },
    "aux_loss_alpha": 0.001,
    "bos_token_id": 163584,
    "dtype": "bfloat16",
    "eos_token_id": 163586,
    "ep_size": 1,
    "first_k_dense_replace": 1,
    "hidden_act": "silu",
    "hidden_size": 7168,
    "initializer_range": 0.02,
    "intermediate_size": 18432,
    "kv_lora_rank": 512,
    "max_position_embeddings": 262144,
    "model_type": "deepseek_v3",
    "moe_intermediate_size": 2048,
    "moe_layer_freq": 1,
    "n_group": 1,
    "n_routed_experts": 384,
    "n_shared_experts": 1,
    "norm_topk_prob": true,
    "num_attention_heads": 64,
    "num_experts_per_tok": 8,
    "num_hidden_layers": 61,
    "num_key_value_heads": 64,
    "num_nextn_predict_layers": 0,
    "pad_token_id": 163839,
    "pretraining_tp": 1,
    "q_lora_rank": 1536,
    "qk_nope_head_dim": 128,
    "qk_rope_head_dim": 64,
    "rms_norm_eps": 1e-05,
    "rope_scaling": {
        "beta_fast": 1.0,
        "beta_slow": 1.0,
        "factor": 64.0,
        "mscale": 1.0,
        "mscale_all_dim": 1.0,
        "original_max_position_embeddings": 4096,
        "type": "yarn"
    },
    "rope_theta": 50000.0,
    "routed_scaling_factor": 2.827,
    "scoring_func": "sigmoid",
    "seq_aux": true,
    "tie_word_embeddings": false,
    "topk_group": 1,
    "topk_method": "noaux_tc",
    "transformers_version": "4.57.1",
    "use_cache": true,
    "v_head_dim": 128,
    "vocab_size": 163840,
    "quantization_config": {
        "global_quant_config": {
            "input_tensors": {
                "dtype": "fp4",
                "is_dynamic": true,
                "qscheme": "per_group",
                "ch_axis": -1,
                "group_size": 32,
                "symmetric": null,
                "round_method": "half_even",
                "scale_type": "float",
                "scale_format": "e8m0",
                "scale_calculation_mode": "even",
                "mx_element_dtype": null,
                "observer_cls": "PerBlockMXObserver",
                "is_scale_quant": false
            },
            "output_tensors": null,
            "weight": {
                "dtype": "fp4",
                "is_dynamic": false,
                "qscheme": "per_group",
                "ch_axis": -1,
                "group_size": 32,
                "symmetric": null,
                "round_method": "half_even",
                "scale_type": "float",
                "scale_format": "e8m0",
                "scale_calculation_mode": "even",
                "mx_element_dtype": null,
                "observer_cls": "PerBlockMXObserver",
                "is_scale_quant": false
            },
            "bias": null,
            "target_device": null
        },
        "exclude": [
            "re:model.layers.*self_attn.*",
            "re:model.layers.*mlp.gate",
            "lm_head",
            "re:model.layers.*mlp.gate_proj",
            "re:model.layers.*mlp.up_proj",
            "re:model.layers.*mlp.down_proj",
            "re:model.layers.*shared_experts.*"
        ],
        "algo_config": null,
        "softmax_quant_spec": null,
        "quant_method": "quark",
        "layer_type_quant_config": {},
        "layer_quant_config": {},
        "kv_cache_quant_config": {},
        "kv_cache_post_rope": false,
        "quant_mode": "eager_mode",
        "version": "0.12+15553c6a61",
        "export": {
            "kv_cache_group": [],
            "min_kv_scale": 0.0,
            "pack_method": "reorder",
            "weight_format": "real_quantized",
            "weight_merge_groups": null
        }
    }
}