Update README.md
Browse files
README.md
CHANGED
|
@@ -15,7 +15,7 @@ Semi-Healed Llama-3 15B Frankenmerge
|
|
| 15 |
|
| 16 |
# Llama-3-Instruct-15B-SPPO-Iter3-SH
|
| 17 |
|
| 18 |
-
|
| 19 |
|
| 20 |
Self-Play Preference Optimization for Language Model Alignment (https://arxiv.org/abs/2405.00675)
|
| 21 |
|
|
|
|
| 15 |
|
| 16 |
# Llama-3-Instruct-15B-SPPO-Iter3-SH
|
| 17 |
|
| 18 |
+
Fully functional upscaled version of [Llama-3-Instruct-8B-SPPO-Iter3](https://huggingface.co/UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3) to 15B parameters with projection swap.
|
| 19 |
|
| 20 |
Self-Play Preference Optimization for Language Model Alignment (https://arxiv.org/abs/2405.00675)
|
| 21 |
|