WizardMath training problem:
ahong007007 opened this issue · comments
ahong007007 commented
Thanks for your work and open source spirit.
WizardMath training problem: The paper claims to use Reinforced Evol-Instruct, but there is no relevant content for code training, which is similar to WizardLM/WizardCoder.
How about the training process Reinforced Evol-Instruct method?
thank you very much!