oskdabk
/

mcqadata_W4A8_0.7

Text Generation

text-generation-inference

compressed-tensors

Model card Files Files and versions

mcqadata_W4A8_0.7

File size: 170 Bytes

a1f7761

default_stage:
  default_modifiers:
    SmoothQuantModifier: {smoothing_strength: 0.7}
    GPTQModifier:
      scheme: W4A8
      targets: Linear
      ignore: [lm_head]