jan-hq/trinity-v1 DPO-trained on Intel/orca_dpo_pairs
#1 Model on the Leaderboard of ANY SIZE 12/16/2023
12/18 Update: Some of the datasets used to create the model I fine-tuned may have been contaminated. I am doing my best to remove thie contamination in future models. Thanks for your patience. Contains traces of Cybertron-2:
title={Cybertron: Uniform Neural Alignment},
author={Xavier Murias},
year={2023},
publisher = {HuggingFace},
journal = {HuggingFace repository},
howpublished = {\url{https://huggingface.co/fblgit/una-cybertron-7b-v2-bf16}},
}```