File size: 180 Bytes
558b191
 
 
3c66597
558b191
 
1
2
3
4
5
6
7
default_stage:
  default_modifiers:
    QuantizationModifier:
      ignore: [lm_head, 're:.*mlp.gate$', 're:.*mlp.shared_expert_gate$']
      targets: [Linear]
      scheme: W4A16