File size: 1,214 Bytes
219f93a 5f225b5 219f93a 3ea2ff0 a94167a 3ea2ff0 af7bbe0 5f225b5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 | ---
license: cc0-1.0
datasets:
- HuggingFaceFW/fineweb-edu
tags:
- gpt
- llm
- lm
- 0.4b
- b
---
# cc0-1.0
[](https://huggingface.co/)
[]()
VDrontV2-0.4b is the smallest version of VDrontV2.
This is *not a chat-based* version! It is designed for learning *basic facts and knowledge, as well as English grammar*.
It is needed for fine-tuning, hence CC0-1.0.
If you’d like to let us know that you’ve improved our model, we’d be delighted—and might even add your version to our collection. However, this is entirely optional: feel free to do whatever you like with it, as the model is released under the CC0 license.
Trained on *FineWeb-Edu, Instruct*!
We are not disclosing the number of training tokens. However, please note that there is still room to reach the theoretical optimum (20 tokens per parameter), making the model suitable for fine-tuning or full retraining.
# To use it, download the repository, create a `model/` folder and place the model files inside it, then place `use.py` (included in the repository) next to that folder and run it. |