Training Script
#2
by
sultan93
- opened
if you can share the fine tuning script using lora, it will very helpful, and also what's your thought on your finetuned model performance .
sultan93
changed discussion status to
closed
@sultan93
for me it was okay, not the best, but you can sometimes get good poems with some playing with the decoding configuration.
You can find the script on the project's repo here: https://github.com/BodaSadalla98/llm-optimized-fintuning