Update README.md
Browse files
README.md
CHANGED
|
@@ -23,7 +23,7 @@ Tags:
|
|
| 23 |
|
| 24 |
**DiscoLM Mixtral 8x7b alpha** is an experimental 8x7b MoE model based on [Mistral AI´s Mixtral 8x7b](https://twitter.com/MistralAI/status/1733150512395038967).
|
| 25 |
This model is based on experimental code converting the model weights to huggingface format and enabling Transformers-based inference.
|
| 26 |
-
It was then finetuned on the
|
| 27 |
DiscoLM Mixtral 8x7b alpha is a [DiscoResearch](https://huggingface.co/DiscoResearch) project and was created by [Björn Plüster](https://huggingface.co/bjoernp) with lots of support from the community.
|
| 28 |
|
| 29 |
**Many thanks to [HessianAI](https://hessian.ai/) for providing the compute resources for this project and to the great people at [LAION](https://laion.ai) without whom this project would not have been possible!**
|
|
|
|
| 23 |
|
| 24 |
**DiscoLM Mixtral 8x7b alpha** is an experimental 8x7b MoE model based on [Mistral AI´s Mixtral 8x7b](https://twitter.com/MistralAI/status/1733150512395038967).
|
| 25 |
This model is based on experimental code converting the model weights to huggingface format and enabling Transformers-based inference.
|
| 26 |
+
It was then finetuned on the Synthia, MethaMathQA und Capybara datasets.
|
| 27 |
DiscoLM Mixtral 8x7b alpha is a [DiscoResearch](https://huggingface.co/DiscoResearch) project and was created by [Björn Plüster](https://huggingface.co/bjoernp) with lots of support from the community.
|
| 28 |
|
| 29 |
**Many thanks to [HessianAI](https://hessian.ai/) for providing the compute resources for this project and to the great people at [LAION](https://laion.ai) without whom this project would not have been possible!**
|