recipe.yaml

quant_stage:
  quant_modifiers:
    SmoothQuantModifier:
      smoothing_strength: 0.7
      mappings:
      - - ['re:.*qkv_proj']
        - re:.*input_layernorm
      - - ['re:.*gate_up_proj']
        - re:.*post_attention_layernorm
    GPTQModifier:
      ignore: [lm_head]
      dampening_frac: 0.09999999999999999
      config_groups:
        group_0:
          targets: [Linear]
          weights: {num_bits: 8, type: int, symmetric: true, strategy: channel, observer: mse}
          input_activations: {num_bits: 8, type: int, symmetric: true, strategy: token, dynamic: true,
            observer: memoryless}