{
  "architectures": [
    "MiMoForCausalLM"
  ],
  "attention_bias": true,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "XiaomiMiMo/MiMo-7B-Base--configuration_mimo.MiMoConfig",
    "AutoModel": "XiaomiMiMo/MiMo-7B-Base--modeling_mimo.MiMoModel",
    "AutoModelForCausalLM": "XiaomiMiMo/MiMo-7B-Base--modeling_mimo.MiMoForCausalLM"
  },
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 32768,
  "max_window_layers": 32,
  "model_type": "mimo",
  "num_attention_heads": 32,
  "num_hidden_layers": 36,
  "num_key_value_heads": 8,
  "num_nextn_predict_layers": 1,
  "quantization_config": {
    "act_bits": 16,
    "act_data_type": "int",
    "act_dynamic": true,
    "act_group_size": 128,
    "act_sym": true,
    "amp": true,
    "autoround_version": "0.5.1",
    "batch_size": 8,
    "bits": 4,
    "damp_percent": 0.01,
    "data_type": "int",
    "desc_act": false,
    "enable_minmax_tuning": true,
    "enable_norm_bias_tuning": false,
    "enable_quanted_input": true,
    "gradient_accumulate_steps": 1,
    "group_size": 128,
    "iters": 1000,
    "low_gpu_mem_usage": true,
    "lr": 0.001,
    "minmax_lr": 0.001,
    "nsamples": 512,
    "quant_method": "gptq",
    "scale_dtype": "torch.float16",
    "seqlen": 2048,
    "super_bits": null,
    "super_group_size": null,
    "sym": true,
    "to_quant_block_names": null,
    "true_sequential": false
  },
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 640000,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.51.3",
  "use_cache": true,
  "use_mrope": false,
  "use_sliding_window": false,
  "vocab_size": 151680
}