winglian/omega-3b

<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

Omega 2.6B

This model is derived from phi 1.3B using layer stacking techniques to double the number of hidden layers in the model. The model was then trained for 1 epoch on data from tiny-textbooks and tiny-lessons.

Training

https://wandb.ai/wing-lian/phi-2x-pt-tiny