---
license: apache-2.0
---

# **Llamion-14B**

We have released Llamion (Llamafied Orion) by transforming [Orion-14B](https://huggingface.co/OrionStarAI/Orion-14B-Base)
into [the standard LLaMA architecture](https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py)
through parameter mapping and offline distillation.
Further technical specifications and study results will be detailed in our upcoming paper, available on this page.

Note that this repository will be deprecated as it is migrated into the [GeM2-Llamion](https://huggingface.co/vaiv/GeM2-Llamion-14B-Base).

![vaiv_png](./vaiv.png)

### Contributors

- VAIV Company AI Lab ([vaiv.kr](https://www.vaiv.kr/))