default_stage:
  default_modifiers:
    GPTQModifier:
      config_groups:
        experts:
          targets: ['re:model.*mlp.*(gate|up|down)_proj$']
          weights:
            num_bits: 4
            type: float
            symmetric: true
            group_size: 16
            strategy: tensor_group
            block_structure: null
            dynamic: false
            actorder: null
            scale_dtype: torch.float8_e4m3fn
            zp_dtype: null
            observer: memoryless_minmax
            observer_kwargs: {}
          input_activations:
            num_bits: 4
            type: float
            symmetric: true
            group_size: 16
            strategy: tensor_group
            block_structure: null
            dynamic: local
            actorder: null
            scale_dtype: torch.float8_e4m3fn
            zp_dtype: null
            observer: static_minmax
            observer_kwargs: {}
          output_activations: null
          format: null
      targets: [Linear]
      ignore: [lm_head, 're:model.*self_attn.*re:model.*ffn_hc$']
      bypass_divisibility_checks: false
      block_size: 128
      dampening_frac: 0.01
      actorder: static
      offload_hessians: false