Pinned Loading

1. sycophancy-eval sycophancy-eval Public

   datasets from the paper "Towards Understanding Sycophancy in Language Models"

   Jupyter Notebook 70 7

2. activation-steering activation-steering Public

   finding a bias-correlated activation vector for llama-2-7b-chat

   Python 3

3. oscarkey/multitask-learning oscarkey/multitask-learning Public

   MSc group project: Reproduction of 'Multi-Task Learning using Uncertainty to Weigh Losses for Scene Geometry and Semantics'; A. Kendall, Y. Gal, R. Cipolla

   Jupyter Notebook 86 12