onkarpandit-g42
commited on
Commit
•
4450759
1
Parent(s):
d9b2917
Update README.md
Browse files
README.md
CHANGED
@@ -188,7 +188,7 @@ The training process was performed on the Condor Galaxy 1 (CG-1) supercomputer p
|
|
188 |
| Precision | fp32 |
|
189 |
| Optimizer | AdamW |
|
190 |
| Learning rate | 0 to 6.7e-04 (<= 400 steps) |
|
191 |
-
| | 6.7e-04 to 6.7e-05 (>
|
192 |
| Weight decay | 0.1 |
|
193 |
| Batch size | 3392 |
|
194 |
| Steps | 8705 |
|
|
|
188 |
| Precision | fp32 |
|
189 |
| Optimizer | AdamW |
|
190 |
| Learning rate | 0 to 6.7e-04 (<= 400 steps) |
|
191 |
+
| | 6.7e-04 to 6.7e-05 (> 400 steps) |
|
192 |
| Weight decay | 0.1 |
|
193 |
| Batch size | 3392 |
|
194 |
| Steps | 8705 |
|