Update README.md
Browse files
README.md
CHANGED
@@ -16,7 +16,7 @@ Note: This model has not been fine-tuned on labeled text data.
|
|
16 |
## Alternative Version
|
17 |
|
18 |
An alternative version of the model which was pre-trained on the same dataset but
|
19 |
-
|
20 |
as a fairseq checkpoint and may give better downstream results.
|
21 |
|
22 |
|
|
|
16 |
## Alternative Version
|
17 |
|
18 |
An alternative version of the model which was pre-trained on the same dataset but
|
19 |
+
with setting `layer_norm_first` to `false` is available [here](https://drive.google.com/file/d/1rbP-6pZfR5ieqAwd5_X2KzipLuKpXSsQ/view?usp=sharing)
|
20 |
as a fairseq checkpoint and may give better downstream results.
|
21 |
|
22 |
|