gowitheflow
commited on
Commit
•
a2ba71d
1
Parent(s):
e50b169
Update README.md
Browse files
README.md
CHANGED
@@ -61,7 +61,7 @@ The model is not for further fine-tuning to do other tasks (such as classificati
|
|
61 |
|
62 |
## Training Details
|
63 |
|
64 |
-
max seq 256, batch size
|
65 |
|
66 |
### Training Data
|
67 |
|
|
|
61 |
|
62 |
## Training Details
|
63 |
|
64 |
+
max seq 256, batch size 128, lr 3e-05, 1 epoch, 10% warmup, 1 A100.
|
65 |
|
66 |
### Training Data
|
67 |
|