Research
Research index
Everything we have published, newest first. Filter by topic, or type to search titles, authors, and abstracts.

All posts
/
6 posts
| Date | Title | Authors | Topic |
|---|---|---|---|
| A public test set we promise not to train on Nadia Haddad, Gabriel Côté, Sorcha Lennox · Policy | Nadia Haddad, Gabriel Côté, Sorcha Lennox | ||
| Why long context fails quietly Rohan Desai, Juliette Bouchard, Ewan Fraser · Reasoning | Rohan Desai, Juliette Bouchard, Ewan Fraser | ||
| Measuring honesty under pressure Isla Mackenzie, Théo Lavoie, Priya Raman · Alignment | Isla Mackenzie, Théo Lavoie, Priya Raman | ||
| Counting with attention: a toy model of tally circuitscoming soon Amara Okafor, Callum Reid · Interpretability | Amara Okafor, Callum Reid | ||
| Sparse features in small vision models Mei-Lin Zhou, Callum Reid, Amara Okafor · Interpretability | Mei-Lin Zhou, Callum Reid, Amara Okafor | ||
| Disclosure norms for dangerous-capability evaluationscoming soon Gabriel Côté, Nadia Haddad · Policy | Gabriel Côté, Nadia Haddad |