|
--- |
|
license: other |
|
--- |
|
|
|
# Join our Discord! https://discord.gg/Nbv9pQ88Xb |
|
|
|
### Works on [Kobold 1.74](https://github.com/LostRuins/koboldcpp/releases/tag/v1.74)! |
|
|
|
*([Layla (iOS / Android)](https://www.layla-network.ai/) support is in progress)* |
|
|
|
--- |
|
|
|
[BeaverAI](https://huggingface.co/BeaverAI) proudly presents... |
|
|
|
# Hubble 4B v1 |
|
|
|
*Equipped with his five senses, man explores the universe around him and calls the adventure 'Science'.* |
|
|
|
 |
|
|
|
## Description |
|
|
|
This is a finetune of Nvidia's Llama 3.1 4B Minitron - a shrunk down model of Llama 3.1 8B 128K. |
|
|
|
### Usage |
|
- ChatML or Text Completion |
|
|
|
### Links |
|
- Original: https://huggingface.co/TheDrummer/Hubble-4B-v1 |
|
- GGUF: https://huggingface.co/TheDrummer/Hubble-4B-v1-GGUF |
|
- Chadquants: https://huggingface.co/bartowski/Hubble-4B-v1-GGUF |
|
|
|
### Technical Note |
|
|
|
Hubble was trained on ChatML with `<|end_of_text|>` as the EOS token, and then swapped to `<|im_end|>` after training (refer to the past commits in `config.json` and `tokenizer_config.json`). If you encounter any issues with the model, please let me know! |
