Tom Aarsen
commited on
Commit
·
b94cf64
1
Parent(s):
5c66f04
Notify users that flash_attn is recommended
Browse files
README.md
CHANGED
|
@@ -35,7 +35,14 @@ The `gte-modernbert` models demonstrates competitive performance in several text
|
|
| 35 |
|
| 36 |
## Usage
|
| 37 |
|
| 38 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
```python
|
| 41 |
# Requires transformers>=4.48.0
|
|
|
|
| 35 |
|
| 36 |
## Usage
|
| 37 |
|
| 38 |
+
> [!TIP]
|
| 39 |
+
> For `transformers` and `sentence-transformers`, if your GPU supports it, the efficient Flash Attention 2 will be used automatically if you have `flash_attn` installed. It is not mandatory.
|
| 40 |
+
>
|
| 41 |
+
> ```bash
|
| 42 |
+
> pip install flash_attn
|
| 43 |
+
> ```
|
| 44 |
+
|
| 45 |
+
Use with `transformers`
|
| 46 |
|
| 47 |
```python
|
| 48 |
# Requires transformers>=4.48.0
|