how to ignore the quantization
#70
by
zzbysd
- opened
the A100 GPU do not support the quantization how could I ignore it
This is supported on transformers main ! You can even run it on colab: https://colab.research.google.com/drive/15DJv6QWgc49MuC7dlNS9ifveXBDjCWO5?usp=sharing
those are for inference, here for training: https://huggingface.co/openai/gpt-oss-20b/discussions/61#6895068120064f17245407a1