File size: 144 Bytes
3a9931b |
1 2 3 4 5 6 7 8 |
quant_stage:
quant_modifiers:
QuantizationModifier:
ignore: [lm_head]
scheme: FP8
observer: mse
targets: [Linear]
|
3a9931b |
1 2 3 4 5 6 7 8 |
quant_stage:
quant_modifiers:
QuantizationModifier:
ignore: [lm_head]
scheme: FP8
observer: mse
targets: [Linear]
|