AI Safety Weekly: Steering Vectors & Input Optimisation
Steering vectors are a relatively simple way of influencing model behaviour. A direction is given to the model’s internal activations for a given concept, and the behaviour shifts accordingly. But such vectors require full access to the model’s internals. Is there a way to automa
Ez az esemény az alábbi nyilvános naptárfolyamok része is, így a látogatók a teljes sorozathoz térhetnek vissza.
Ido
10 Nov 2026 17:30 – 10 Nov 2026 19:00
Helyszin
ETH HG Hauptgebäude, Zurich
ETH HG Hauptgebäude, Zurich, ch
Kozzetevo
Zurich AI Safety
Az eseményt tartalmazó nyilvános naptárak
Duration
1 óra 30 perc
Responsible
Zurich AI Safety
Source details
- Original event
- https://luma.com/zais-42em
- Import status
- Regional Scan
- Source kind
- Natív
- Updated
- 1 napja
- Public calendars
- Switzerland Tech Events
Eseménytérkép
ETH HG Hauptgebäude, Zurich, ch
Térképadatok © OpenStreetMap közreműködők
Steering vectors are a relatively simple way of influencing model behaviour. A direction is given to the model’s internal activations for a given concept, and the behaviour shifts accordingly. But such vectors require full access to the model’s internals. Is there a way to automatically optimise an input to induce the same behaviour as a steering vector? Can this input be transferred to different Price / conditions: free