File size: 186 Bytes
d386232
 
 
 
 
 
1
2
3
4
5
6
7
default_stage:
  default_modifiers:
    QuantizationModifier:
      targets: [Linear]
      ignore: [lm_head, 're:.*mlp.gate$', 're:.*mlp.shared_expert_gate$']
      scheme: FP8_dynamic