README.md

metadata

library_name: transformers
license: cc-by-nc-4.0
tags:
  - merge
  - automerger

UltraMerge-7B

This model is an experimental DPO fine-tune of automerger/YamShadow-7B on the following datasets:

• mlabonne/truthy-dpo-v0.1
• mlabonne/distilabel-intel-orca-dpo-pairs
• mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha
• mlabonne/ultrafeedback-binarized-preferences-cleaned

I have no idea about what's the best chat template. Probably Mistral-Instruct or ChatML.