Update README.md
Browse files
README.md
CHANGED
|
@@ -80,7 +80,6 @@ If you want to scale your model context by 2x:
|
|
| 80 |
```python
|
| 81 |
model = AutoModel.from_pretrained(model_name,
|
| 82 |
trust_remote_code=True,
|
| 83 |
-
attn_implementation='sdpa',
|
| 84 |
rope_scaling={'type': 'dynamic','factor': 2.0}
|
| 85 |
) # 2.0 for x2 scaling, 4.0 for x4, etc..
|
| 86 |
```
|
|
|
|
| 80 |
```python
|
| 81 |
model = AutoModel.from_pretrained(model_name,
|
| 82 |
trust_remote_code=True,
|
|
|
|
| 83 |
rope_scaling={'type': 'dynamic','factor': 2.0}
|
| 84 |
) # 2.0 for x2 scaling, 4.0 for x4, etc..
|
| 85 |
```
|