Interpretability
Opening the black box: reading a model's internal features.
Stai visualizzando una scheda concettuale: cosa significa questo termine in linguaggio chiaro, e i record che mette in relazione.
Mechanistic interpretability reverse-engineers what networks compute — Anthropic's team mapped millions of human-legible features in Claude (the 'Golden Gate Claude' demo made one famous) and traces circuits for planning and deception. The bet: we cannot fully trust what we cannot inspect, so understanding must catch up with capability before the stakes get too high.
Come si collega al resto dello stack
I collegamenti di questa scheda a colpo d'occhio — clicca un gruppo per raggiungerlo.
Correlati
Fonti
Mapping the mind of an LLM (Anthropic) ↗
Record revisionato al 2026-01-15 — vedi Metodologia.