Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -3,6 +3,9 @@ library_name: transformers
|
|
| 3 |
tags: []
|
| 4 |
---
|
| 5 |
|
|
|
|
|
|
|
|
|
|
| 6 |
# FastESM
|
| 7 |
FastESM is a Huggingface compatible plug in version of ESM2 rewritten with a newer PyTorch attention implementation.
|
| 8 |
|
|
|
|
| 3 |
tags: []
|
| 4 |
---
|
| 5 |
|
| 6 |
+
# NOTE
|
| 7 |
+
There was previously a bug with Huggingface weight tieing that caused the logits of FastESM to differ from ESM2. That bug is now resolved.
|
| 8 |
+
|
| 9 |
# FastESM
|
| 10 |
FastESM is a Huggingface compatible plug in version of ESM2 rewritten with a newer PyTorch attention implementation.
|
| 11 |
|