Files
tinyllama-oneshot-w8a16-per…/recipe.yaml
ModelHub XC 95cf8e4b82 初始化项目,由ModelHub XC社区提供模型
Model: nm-testing/tinyllama-oneshot-w8a16-per-channel
Source: Original Platform
2026-10-23 02:33:13 +08:00

10 lines
251 B
YAML

quant_stage:
quant_modifiers:
GPTQModifier:
sequential_update: false
ignore: [lm_head]
config_groups:
group_0:
weights: {num_bits: 8, type: int, symmetric: true, strategy: channel}
targets: [Linear]