Update README.md
Browse files
README.md
CHANGED
|
@@ -68,7 +68,6 @@ We trained `google/umt5-small` [300 million parameters (~1.20 GB)] on the GreekS
|
|
| 68 |
* Total training epochs = 10
|
| 69 |
* AdamW optimizer (e = 1e−8, β1 = 0.9 and β2 = 0.0999)
|
| 70 |
* Learning rate = 3e−4
|
| 71 |
-
* Linear weight decay
|
| 72 |
* No warmup steps
|
| 73 |
* 32-bit floating precision
|
| 74 |
* Tokenization
|
|
|
|
| 68 |
* Total training epochs = 10
|
| 69 |
* AdamW optimizer (e = 1e−8, β1 = 0.9 and β2 = 0.0999)
|
| 70 |
* Learning rate = 3e−4
|
|
|
|
| 71 |
* No warmup steps
|
| 72 |
* 32-bit floating precision
|
| 73 |
* Tokenization
|