mamba-2.8b-slimorca / README.md
rwl4's picture
Upload folder using huggingface_hub
74074f9
|
raw
history blame
No virus
678 Bytes
metadata
license: apache-2.0

Mamba-2.8B-SlimOrca Model

Overview

Mamba-2.8B-SlimOrca is a finetune of the original pretrained Mamba model, utilizing the SlimOrca dataset. At this point, it's still pretty rough.

Usage

The easiest way to try out this model is to use chat.py from Mamba-Chat.

Future Plans

  • Integration with Hugging Face space (planned). Contributions welcome.

Model Training Details

  • Dataset Used: SlimOrca
  • Training Duration: Single epoch
  • Current Status: Training is ongoing. Further iterations will be uploaded if there are improvements.