---
datasets:
  - ewof/koishi-instruct-metharme
  - PygmalionAI/PIPPA
---

just testing for now, qlora marge, several things different between this and the 7b

https://rentry.org/v43eo updated the rentry slightly

![image/png](https://cdn-uploads.huggingface.co/production/uploads/64f38f3a89203d118da8b477/HkWCy5CQwlBGEpmmLAwuO.png)

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