wanda icon indicating copy to clipboard operation
wanda copied to clipboard
verified publisher icon locuslab

Results After LoRA Fine-Tuning

Open hhhhain opened this issue 1 year ago • 0 comments

The perplexity of the LLAMA-7B model fine-tuned using the 'script.sh' from the 'lora_ft' file differs significantly from the results presented in your paper. Could you kindly advise if there might be an issue with the fine-tuning code?

hhhhain avatar Aug 13 '24 10:08 hhhhain