Link paper (arXiv:2611.05287)
Browse files
README.md
CHANGED
|
@@ -31,3 +31,5 @@ The training data and statements are in [`knowingly-wrong/realistic-data`](https
|
|
| 31 |
from peft import PeftModel
|
| 32 |
model = PeftModel.from_pretrained(base, "knowingly-wrong/realistic-scenarios", subfolder="brexit/instructed_false")
|
| 33 |
```
|
|
|
|
|
|
|
|
|
| 31 |
from peft import PeftModel
|
| 32 |
model = PeftModel.from_pretrained(base, "knowingly-wrong/realistic-scenarios", subfolder="brexit/instructed_false")
|
| 33 |
```
|
| 34 |
+
|
| 35 |
+
Paper: [Knowingly Wrong: Training LLMs Against Their Own Beliefs can Cause Misalignment](https://arxiv.org/abs/2611.05287)
|