Publish final model documentation with measured results and limitations
Browse files
README.md
CHANGED
|
@@ -313,3 +313,9 @@ These experiments change parameter counts, representations, training signal, bud
|
|
| 313 |
## License
|
| 314 |
|
| 315 |
The model repository declares Apache-2.0. Upstream datasets, corpus excerpts and tokenizer assets retain their own licenses.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 313 |
## License
|
| 314 |
|
| 315 |
The model repository declares Apache-2.0. Upstream datasets, corpus excerpts and tokenizer assets retain their own licenses.
|
| 316 |
+
|
| 317 |
+
## Related resources
|
| 318 |
+
|
| 319 |
+
- [GitHub: igna-s/New-1.58](https://github.com/igna-s/New-1.58): extraction pipeline, training and evaluation code.
|
| 320 |
+
- [Hugging Face: Cukinator/cpu1-ablation-checkpoints](https://huggingface.co/Cukinator/cpu1-ablation-checkpoints): packed and intermediate checkpoints.
|
| 321 |
+
- [Hugging Face: Cukinator/cpu1-ablation-dataset](https://huggingface.co/datasets/Cukinator/cpu1-ablation-dataset): byte and BPE training data.
|