Add paper reference (arXiv:2605.26045) to README body
Browse files
README.md
CHANGED
|
@@ -17,3 +17,7 @@ This adapter is intended to be used in experiments assessing representation engi
|
|
| 17 |
|
| 18 |
## Training Data
|
| 19 |
The model was trained on a split of the `bcywinski/taboo-flame` dataset alongside general chat data (`HuggingFaceH4/ultrachat_200k`) to maintain conversational ability while enforcing the taboo constraint.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Training Data
|
| 19 |
The model was trained on a split of the `bcywinski/taboo-flame` dataset alongside general chat data (`HuggingFaceH4/ultrachat_200k`) to maintain conversational ability while enforcing the taboo constraint.
|
| 20 |
+
|
| 21 |
+
## Related Paper
|
| 22 |
+
|
| 23 |
+
This adapter is one of the taboo target models used in [Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals](https://arxiv.org/abs/2605.26045) (arXiv:2605.26045).
|