Commit
·
8b85a80
1
Parent(s):
3baa823
Update README.md
Browse files
README.md
CHANGED
|
@@ -40,7 +40,7 @@ However, I could not efficiently optimize the second feedforward network sub-lay
|
|
| 40 |
|
| 41 |
*On the left is the standard Series Attention and Feed-Forward Net Design (SAF) for transformers models. On the right is the Parallel Attention and Feed-Forward Net Design (PAF) used in transformer models like PaLM (Chowdhery et al., 2022) and Mesh-Transformers (Wang, 2021)*
|
| 42 |
|
| 43 |
-
## Evaluation results of [PAF-RoBERTa-Large](https://huggingface.co/luffycodes/parallel-roberta-large)
|
| 44 |
|
| 45 |
When fine-tuned on downstream tasks, this model achieves the following results:
|
| 46 |
|
|
|
|
| 40 |
|
| 41 |
*On the left is the standard Series Attention and Feed-Forward Net Design (SAF) for transformers models. On the right is the Parallel Attention and Feed-Forward Net Design (PAF) used in transformer models like PaLM (Chowdhery et al., 2022) and Mesh-Transformers (Wang, 2021)*
|
| 42 |
|
| 43 |
+
## Evaluation results of [PAF-RoBERTa-Large](https://huggingface.co/luffycodes/parallel-roberta-large)
|
| 44 |
|
| 45 |
When fine-tuned on downstream tasks, this model achieves the following results:
|
| 46 |
|