Quantization made by Richard Erkhov.
NSFW_DPO_vmgb-7b - bnb 4bits
Original model description:
license: cc-by-nc-4.0
datasets:
- athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v2
- athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW
language:
- en
v1olet/v1olet_marcoroni-go-bruins-merge-7B trained for an epoch on my NSFW_DPO-v1 dataset, then the some LoRA state was trained until crash on DPO-v2 dataset (made private until I can figure it out), then again from that point on 1 more epoch of the NSFW_DPO-v1 dataset
