viren-shah
commited on
Commit
•
e8cc5ce
1
Parent(s):
ff9b04c
Update README.md
Browse files
README.md
CHANGED
@@ -126,7 +126,6 @@ We applied instruction tuning on a variety of tasks derived from datasets such a
|
|
126 |
- Steps: 35000
|
127 |
- Global Batch size: 64
|
128 |
- Learning Rate: 1e-5
|
129 |
-
- Learning Rate: 1e-5
|
130 |
- Learning Rate Scheduler: Fixed
|
131 |
- Warmup Steps: 0
|
132 |
- Weight decay: 0.1
|
|
|
126 |
- Steps: 35000
|
127 |
- Global Batch size: 64
|
128 |
- Learning Rate: 1e-5
|
|
|
129 |
- Learning Rate Scheduler: Fixed
|
130 |
- Warmup Steps: 0
|
131 |
- Weight decay: 0.1
|