Safetensors
gpt2

cc0-1.0

Hugging Face Parameters

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.

Downloads last month
39
Safetensors
Model size
0.4B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for VDC-team/VDrontV2-0.4b

Quantizations
1 model

Collection including VDC-team/VDrontV2-0.4b