Iambe-RP-cDPO-20b-ALT

<p align="center"><img src="https://i.ibb.co/PCY2Xym/Iambe-RP-sml-t.png"/><font size="6"> <b>Iambe-RP-cDPO-20b-ALT</b> </font></p> <p align="center"><font size="4"> <b>Trained with Alpaca prompt formatting, some like ChatML</b> </font></p>

Description

Named after a charming daughter of Echo and Pan in Greek myth, Iambe-RP-ALT failed training on the v2 dataset at 0.6 epochs. Therefore, I don't feel comfortable considering this a full new generation.

Iambe is intended to have the best realistically possible understanding of instructions, anatomy and scene state for a 20b merge, while remaining passionate and humanoid in "voice".

Update Methodology

Take a look at the dataset v2 Iambe and I created together for more info. The cDPO training was done directly on Iambe-Storyteller-20b, and I The notebook used is also available in the dataset's repo.

<p align="center"><font size="5"> <b>Assistant Example @ q5_k_m</b> </font></p> <p align="center"><img src="https://i.ibb.co/YtwztYH/Screenshot-2023-12-09-113216.png"/> <p align="center"><font size="5"> <b>NSFW Writing Example @ q5_k_m</b> </font></p> <p align="center"><img src="https://i.ibb.co/4JHGzph/Screenshot-2023-12-09-113534.png"/> <p align="center"><font size="5"> <b>Silly but Impressive RP Example @ q5_k_m</b> </font></p> <p align="center"><img src="https://i.ibb.co/XX0FDkx/Screenshot-2023-12-09-115015.png"/>