|
--- |
|
language: fr |
|
license: mit |
|
tags: |
|
- pytorch |
|
- causal-lm |
|
datasets: |
|
- c4 |
|
--- |
|
|
|
# Cedille AI |
|
Cedille is a project to bring large language models to non-English languages. |
|
|
|
## Boris |
|
Boris is a 6B parameter autoregressive language model based on the GPT-J architecture and trained using the (mesh-transformer-jax)[https://github.com/kingoflolz/mesh-transformer-jax] codebase. |
|
|
|
Boris was trained on around 78B tokens of French text from the [C4](https://huggingface.co/datasets/c4) dataset. We started training from GPT-J, which has been trained on [The Pile](https://pile.eleuther.ai/). As a consequence the model still has good performance in English language. Boris makes use of the unmodified GPT-2 tokenizer. |
|
|
|
Boris is named after the great French writer [Boris Vian](https://en.wikipedia.org/wiki/Boris_Vian). |
|
|
|
# How do I test Cedille? |
|
For the time being, the easiest way to test the model is to use our [publicly accessible playground](https://en.cedille.ai/). |
|
|
|
Cedille is a relatively large model and running it in production can get expensive. Consider contacting us for API access. |
|
|
|
# How do I cite Cedille? |
|
Thanks for citing our work in case you build on top of Cedille. For the time being, please reference our work like so: |
|
```bibtex |
|
@misc{cedille2021, |
|
author = {Laurent, F. and Muller, M.}, |
|
title = {Cedille AI}, |
|
year = {2021}, |
|
publisher = {GitHub}, |
|
journal = {GitHub repository}, |
|
howpublished = {\url{https://github.com/coteries/cedille-ai}} |
|
} |
|
``` |
|
|
|
## Contact us |
|
For any custom development please contact us at info@coteries.com. |
|
|
|
## Links |
|
* [Official website](https://en.cedille.ai/) |
|
* [Blog](https://en.cedille.ai/blog) |
|
* [GitHub](https://github.com/coteries/cedille-ai) |
|
* [Twitter](https://twitter.com/CedilleAI) |
|
|
|
|