default_stage:
  default_modifiers:
    AWQModifier:
      config_groups:
        mlp_experts_projections:
          targets: ['re:.*block_sparse_moe\.experts\.[0-9]+\.(w1|w2|w3)$']
          weights:
            num_bits: 4
            type: int
            symmetric: true
            group_size: 32
            strategy: group
            block_structure: null
            dynamic: false
            actorder: null
            scale_dtype: null
            zp_dtype: null
            observer: minmax
            observer_kwargs: {}
          input_activations: null
          output_activations: null
          format: null
      targets: [Linear]
      ignore: [lm_head, embed_tokens, 're:.*self_attn.*', 're:.*block_sparse_moe\.gate$',
        're:.*\.layers\.(58|59|60|61)\.block_sparse_moe\.experts\.[0-9]+\.(w1|w2|w3)$']
      bypass_divisibility_checks: false
      mappings:
      - smooth_layer: re:.*post_attention_layernorm$
        balance_layers: ['re:.*w1$', 're:.*w3$']
        activation_hook_target: null
      - smooth_layer: re:.*w3$
        balance_layers: ['re:.*w2$']
        activation_hook_target: null
      offload_device: !!python/object/apply:torch.device [cpu]
      duo_scaling: true
      n_grid: 20