File size: 231 Bytes
73ebc9b |
1 2 3 4 5 6 7 8 9 |
test_stage:
quant_modifiers:
vLLMQuantizationModifier:
ignore: [lm_head]
config_groups:
group_0:
weights: {num_bits: 4, type: int, symmetric: true, strategy: channel}
targets: [Linear]
|