Add proper model card with research links
Browse files
README.md
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- bitnet
|
| 4 |
+
- tool-calling
|
| 5 |
+
- training-recipe
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# BitNet tool-calling — training recipe
|
| 9 |
+
|
| 10 |
+
Training script and configuration for fine-tuning a BitNet-architecture model on
|
| 11 |
+
[Adiuk/adis-tool-calling-dataset](https://huggingface.co/datasets/Adiuk/adis-tool-calling-dataset).
|
| 12 |
+
|
| 13 |
+
**This repo contains the recipe (`train.py` + `training_config.json`), not model weights.**
|
| 14 |
+
Part of the Eyla project's exploration of extremely low-bit on-device agent models.
|
| 15 |
+
|
| 16 |
+
## Author & research
|
| 17 |
+
|
| 18 |
+
- 🌐 Website: [arifadito.com](https://arifadito.com)
|
| 19 |
+
- 💻 GitHub: [github.com/Adiuk24](https://github.com/Adiuk24)
|
| 20 |
+
- 📄 Eyla identity paper: [arXiv:2604.00009](https://arxiv.org/abs/2604.00009)
|
| 21 |
+
- 🔧 Rust ML training verification tooling: [gradient-flow-arbiter](https://github.com/Adiuk24/gradient-flow-arbiter)
|