File size: 202 Bytes
ee5fb4d |
1 2 3 4 5 6 7 |
default_stage:
default_modifiers:
QuantizationModifier:
targets: [Linear]
ignore: [lm_head, 're:.*mlp.gate$', 're:.*mlp.shared_expert_gate$', 're:.*linear_attn.*']
scheme: NVFP4
|