Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model:
|
| 3 |
+
- allura-org/Koto-22B-PT
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
Testish instruct checkpoint of Koto.
|
| 7 |
+
|
| 8 |
+
### Recipe:
|
| 9 |
+
Upscale Nemo 12B to 22B
|
| 10 |
+
|
| 11 |
+
Do a continual pretrain for 1B tokens on creative data (big thanks to Allura Org for doing this)
|
| 12 |
+
|
| 13 |
+
Train on ~ 17K instruct samples.
|
| 14 |
+
|
| 15 |
+
Uses alpaca format.
|