metadata
datasets:
- ewof/koishi-instruct-metharme
- PygmalionAI/PIPPA
just testing for now, qlora marge
https://rentry.org/v43eo updated the rentry slightly
NousResearch/Llama-2-13b-hf tuned on koishi data (without code subsets) for 1 epoch
then tuned on pippa for 1 epoch
then tuned on limarp (without ponyville, lolicit, all the fallen, and eka's portal subsets) for 2 epochs
all metharme format