quantization_config.json

{
  "bits": 4,
  "group_size": 128,
  "sym": true,
  "data_type": "int",
  "nsamples": 512,
  "autoround_version": "0.6.1.dev",
  "quant_method": "auto-round",
  "packing_format": "auto_round:auto_gptq",
  "extra_config": {
    "model.layers.0.mlp.gate": {
      "bits": 16
    },
    "model.layers.1.mlp.gate": {
      "bits": 16
    },
    "model.layers.2.mlp.gate": {
      "bits": 16
    },
    "model.layers.3.mlp.gate": {
      "bits": 16
    },
    "model.layers.4.mlp.gate": {
      "bits": 16
    },
    "model.layers.5.mlp.gate": {
      "bits": 16
    },
    "model.layers.6.mlp.gate": {
      "bits": 16
    },
    "model.layers.7.mlp.gate": {
      "bits": 16
    },
    "model.layers.8.mlp.gate": {
      "bits": 16
    },
    "model.layers.9.mlp.gate": {
      "bits": 16
    },
    "model.layers.10.mlp.gate": {
      "bits": 16
    },
    "model.layers.11.mlp.gate": {
      "bits": 16
    },
    "model.layers.12.mlp.gate": {
      "bits": 16
    },
    "model.layers.13.mlp.gate": {
      "bits": 16
    },
    "model.layers.14.mlp.gate": {
      "bits": 16
    },
    "model.layers.15.mlp.gate": {
      "bits": 16
    },
    "model.layers.16.mlp.gate": {
      "bits": 16
    },
    "model.layers.17.mlp.gate": {
      "bits": 16
    },
    "model.layers.18.mlp.gate": {
      "bits": 16
    },
    "model.layers.19.mlp.gate": {
      "bits": 16
    },
    "model.layers.20.mlp.gate": {
      "bits": 16
    },
    "model.layers.21.mlp.gate": {
      "bits": 16
    },
    "model.layers.22.mlp.gate": {
      "bits": 16
    },
    "model.layers.23.mlp.gate": {
      "bits": 16
    },
    "model.layers.24.mlp.gate": {
      "bits": 16
    },
    "model.layers.25.mlp.gate": {
      "bits": 16
    },
    "model.layers.26.mlp.gate": {
      "bits": 16
    },
    "model.layers.27.mlp.gate": {
      "bits": 16
    },
    "model.layers.28.mlp.gate": {
      "bits": 16
    },
    "model.layers.29.mlp.gate": {
      "bits": 16
    },
    "model.layers.30.mlp.gate": {
      "bits": 16
    },
    "model.layers.31.mlp.gate": {
      "bits": 16
    },
    "model.layers.32.mlp.gate": {
      "bits": 16
    },
    "model.layers.33.mlp.gate": {
      "bits": 16
    },
    "model.layers.34.mlp.gate": {
      "bits": 16
    },
    "model.layers.35.mlp.gate": {
      "bits": 16
    },
    "model.layers.36.mlp.gate": {
      "bits": 16
    },
    "model.layers.37.mlp.gate": {
      "bits": 16
    },
    "model.layers.38.mlp.gate": {
      "bits": 16
    },
    "model.layers.39.mlp.gate": {
      "bits": 16
    },
    "model.layers.40.mlp.gate": {
      "bits": 16
    },
    "model.layers.41.mlp.gate": {
      "bits": 16
    },
    "model.layers.42.mlp.gate": {
      "bits": 16
    },
    "model.layers.43.mlp.gate": {
      "bits": 16
    },
    "model.layers.44.mlp.gate": {
      "bits": 16
    },
    "model.layers.45.mlp.gate": {
      "bits": 16
    },
    "model.layers.46.mlp.gate": {
      "bits": 16
    },
    "model.layers.47.mlp.gate": {
      "bits": 16
    },
    "model.layers.48.mlp.gate": {
      "bits": 16
    },
    "model.layers.49.mlp.gate": {
      "bits": 16
    },
    "model.layers.50.mlp.gate": {
      "bits": 16
    },
    "model.layers.51.mlp.gate": {
      "bits": 16
    },
    "model.layers.52.mlp.gate": {
      "bits": 16
    },
    "model.layers.53.mlp.gate": {
      "bits": 16
    },
    "model.layers.54.mlp.gate": {
      "bits": 16
    },
    "model.layers.55.mlp.gate": {
      "bits": 16
    },
    "model.layers.56.mlp.gate": {
      "bits": 16
    },
    "model.layers.57.mlp.gate": {
      "bits": 16
    },
    "model.layers.58.mlp.gate": {
      "bits": 16
    },
    "model.layers.59.mlp.gate": {
      "bits": 16
    },
    "model.layers.60.mlp.gate": {
      "bits": 16
    },
    "model.layers.61.mlp.gate": {
      "bits": 16
    },
    "model.layers.62.mlp.gate": {
      "bits": 16
    },
    "model.layers.63.mlp.gate": {
      "bits": 16
    },
    "model.layers.64.mlp.gate": {
      "bits": 16
    },
    "model.layers.65.mlp.gate": {
      "bits": 16
    },
    "model.layers.66.mlp.gate": {
      "bits": 16
    },
    "model.layers.67.mlp.gate": {
      "bits": 16
    },
    "model.layers.68.mlp.gate": {
      "bits": 16
    },
    "model.layers.69.mlp.gate": {
      "bits": 16
    },
    "model.layers.70.mlp.gate": {
      "bits": 16
    },
    "model.layers.71.mlp.gate": {
      "bits": 16
    },
    "model.layers.72.mlp.gate": {
      "bits": 16
    },
    "model.layers.73.mlp.gate": {
      "bits": 16
    },
    "model.layers.74.mlp.gate": {
      "bits": 16
    },
    "model.layers.75.mlp.gate": {
      "bits": 16
    },
    "model.layers.76.mlp.gate": {
      "bits": 16
    },
    "model.layers.77.mlp.gate": {
      "bits": 16
    },
    "model.layers.78.mlp.gate": {
      "bits": 16
    },
    "model.layers.79.mlp.gate": {
      "bits": 16
    },
    "model.layers.80.mlp.gate": {
      "bits": 16
    },
    "model.layers.81.mlp.gate": {
      "bits": 16
    },
    "model.layers.82.mlp.gate": {
      "bits": 16
    },
    "model.layers.83.mlp.gate": {
      "bits": 16
    },
    "model.layers.84.mlp.gate": {
      "bits": 16
    },
    "model.layers.85.mlp.gate": {
      "bits": 16
    },
    "model.layers.86.mlp.gate": {
      "bits": 16
    },
    "model.layers.87.mlp.gate": {
      "bits": 16
    },
    "model.layers.88.mlp.gate": {
      "bits": 16
    },
    "model.layers.89.mlp.gate": {
      "bits": 16
    },
    "model.layers.90.mlp.gate": {
      "bits": 16
    },
    "model.layers.91.mlp.gate": {
      "bits": 16
    },
    "model.layers.92.mlp.gate": {
      "bits": 16
    },
    "model.layers.93.mlp.gate": {
      "bits": 16
    }
  }
}