Thank you for your great effort to release the data and model. I'd like to see the detailed performance on popular benchmark ( especially comparison with origin Qwen 72B Chat Model) to see the power of continue SFT Great appreciate if you can show the results!
