Update README.md
Browse files
README.md
CHANGED
|
@@ -160,7 +160,6 @@ The number of examples per task was capped to 64. The model was trained for 20k
|
|
| 160 |
|
| 161 |
You can fine-tune this model to use it for multiple-choice or any classification task (e.g. NLI) like any debertav2 model.
|
| 162 |
This model has strong zero-shot validation performance on many tasks (e.g. 70% on WNLI).
|
| 163 |
-
|
| 164 |
The untuned model CLS embedding also has strong linear probing performance (90% on MNLI), due to the multitask training.
|
| 165 |
|
| 166 |
The list of tasks is available in tasks.md
|
|
|
|
| 160 |
|
| 161 |
You can fine-tune this model to use it for multiple-choice or any classification task (e.g. NLI) like any debertav2 model.
|
| 162 |
This model has strong zero-shot validation performance on many tasks (e.g. 70% on WNLI).
|
|
|
|
| 163 |
The untuned model CLS embedding also has strong linear probing performance (90% on MNLI), due to the multitask training.
|
| 164 |
|
| 165 |
The list of tasks is available in tasks.md
|