Files
gemma-2-9b-it-quantized.w8a16/recipe.yaml
ModelHub XC e5925ed65f 初始化项目,由ModelHub XC社区提供模型
Model: neuralmagic/gemma-2-9b-it-quantized.w8a16
Source: Original Platform
2026-10-23 07:01:13 +08:00

9 lines
174 B
YAML

quant_stage:
quant_modifiers:
GPTQModifier:
sequential_update: false
dampening_frac: 0.01
ignore: [lm_head]
scheme: W8A16
targets: Linear