File size: 252 Bytes
c1500cf |
1 2 3 4 5 6 7 8 9 10 |
quant_stage:
quant_modifiers:
GPTQModifier:
ignore: [lm_head]
config_groups:
group_0:
weights: {num_bits: 8, strategy: tensor}
input_activations: {num_bits: 8, strategy: tensor}
targets: [Linear]
|