config.json

{
  "architectures": [
    "Qwen3MoeForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "decoder_sparse_step": 1,
  "eos_token_id": 151645,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5472,
  "max_position_embeddings": 262144,
  "max_window_layers": 28,
  "mlp_only_layers": [],
  "model_type": "qwen3_moe",
  "moe_intermediate_size": 768,
  "norm_topk_prob": true,
  "num_attention_heads": 32,
  "num_experts": 128,
  "num_experts_per_tok": 8,
  "num_hidden_layers": 48,
  "num_key_value_heads": 4,
  "output_router_logits": false,
  "qkv_bias": false,
  "quantization_config": {
    "bits": 8,
    "checkpoint_format": "gptq",
    "desc_act": false,
    "group_size": 32,
    "lm_head": false,
    "memory_aware_allocation": false,
    "memory_threshold": 0.8,
    "meta": {
      "damp_auto_increment": 0.01,
      "damp_percent": 0.1,
      "mse": 0.0,
      "quantizer": [
        "gptqmodel:4.0.0-dev"
      ],
      "static_groups": false,
      "true_sequential": true,
      "uri": "https://github.com/modelcloud/gptqmodel",
      "v2": false,
      "v2_alpha": 0.25
    },
    "pack_dtype": "int32",
    "quant_method": "gptq",
    "sym": true
  },
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000000,
  "router_aux_loss_coef": 0.0,
  "shared_expert_intermediate_size": 0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.52.4",
  "use_cache": true,
  "use_qk_norm": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}