Commit
•
4aa61c1
1
Parent(s):
bdb6c63
Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,8 @@ https://huggingface.co/Josephgflowers/Differential-Attention-Liquid-Metal-Tinyll
|
|
| 17 |
|
| 18 |
Continued training for healing consisted of around 58860 steps full training on common datasets like open orca, ultrachat, and textbooks are all you need style datasets.
|
| 19 |
|
|
|
|
|
|
|
| 20 |
|
| 21 |
|
| 22 |
|
|
|
|
| 17 |
|
| 18 |
Continued training for healing consisted of around 58860 steps full training on common datasets like open orca, ultrachat, and textbooks are all you need style datasets.
|
| 19 |
|
| 20 |
+
Benchmarks put this back around the base model for performance, this model could use further continued training or training on downstream tasks.
|
| 21 |
+
|
| 22 |
|
| 23 |
|
| 24 |
|