Analyze Feature Flow to Enhance Interpretation and Steering in Language Models Paper • 2502.03032 • Published 10 days ago • 53
The Differences Between Direct Alignment Algorithms are a Blur Paper • 2502.01237 • Published 12 days ago • 112
Ru Dialogue Benchmarks Collection A collection of benchmarks for evaluating the quality of dialogue models in Russian. • 3 items • Updated Jan 15 • 2