Pinned Loading

1. llm-censorship-steering llm-censorship-steering Public

   Code for 'Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control'

   Python 7

2. gender-bias-steering gender-bias-steering Public

   Python 1

3. automatic-paraphrase-dataset-augmentation automatic-paraphrase-dataset-augmentation Public

   Code and data for automatic paraphrase dataset augmentation.

   Jupyter Notebook 11

4. balanced-adversarial-training balanced-adversarial-training Public

   Python 2

5. composed-debiasing composed-debiasing Public

   Jupyter Notebook 1

6. allocational-harm-eval allocational-harm-eval Public

   Code for evaluating allocational harms in machine learning and large language models.

   Python