mkartofel
/

MNLP_M2_quantized_model

8-bit precision

compressed-tensors

Model card Files Files and versions

MNLP_M2_quantized_model

File size: 170 Bytes

d985b58

default_stage:
  default_modifiers:
    SmoothQuantModifier: {smoothing_strength: 0.8}
    GPTQModifier:
      scheme: W8A8
      targets: Linear
      ignore: [lm_head]