🌸 This is a collection of synthetic datasets built to help improve the ability of open language models to better write haikus through the use of DPO • 3 items • Updated • 6
Model Card for HaikuHermes-0.1-7B
This is a very early model which uses the davanstrien/haiku_dpo dataset to train teknium/OpenHermes-2.5-Mistral-7B using Direct Preference Optimization.
The eventual goal of this model is for it to write "technically correct" haiku.
- Downloads last month
- 6
Safetensors
Model size
7B params
Tensor type
F16
·
Model tree for davanstrien/HaikuHermes-0.1-7B
Base model
mistralai/Mistral-7B-v0.1 Finetuned
teknium/OpenHermes-2.5-Mistral-7B