Commit
•
d1936ec
1
Parent(s):
72da9b3
Update README.md
Browse files
README.md
CHANGED
|
@@ -8,7 +8,9 @@ Based on GPT-Neo architecture.
|
|
| 8 |
|
| 9 |
License: mit
|
| 10 |
|
| 11 |
-
|
|
|
|
|
|
|
| 12 |
lr = 5e-4
|
| 13 |
lr_schedule = constant
|
| 14 |
wd=0.1
|
|
|
|
| 8 |
|
| 9 |
License: mit
|
| 10 |
|
| 11 |
+
---
|
| 12 |
+
hyperparams used to train this model:
|
| 13 |
+
|
| 14 |
lr = 5e-4
|
| 15 |
lr_schedule = constant
|
| 16 |
wd=0.1
|