I keep getting this warning message when I do dpo :The following columns in the training set don't have a corresponding argument in PeftModelForCausalLM.forward and have been ignored: prompt, chosen, rejected. If prompt, chosen, rejected are not expected by PeftModelForCausalLM.forward, you can safely ignore this message. Dpo seems to work fine though but I want to be sure because I am going to invest a lot of time on that.
#The following columns in the training set don't have a corresponding argument in PeftModelForCausalL
5 messages · Page 1 of 1 (latest)
I don't think I got any of that error when I use DPO for my project .-.
is there any chance you can share your code with me? I did many runs with different configurations and always get it. However if I do training directly without unsloth I dont get it anymore.