The letter

One letter a week, starting soon.

The week's signal, what shipped, what is worth running tonight, and the editor's note. No tracking, no ads, nothing else.

We keep your address, your language and the date you joined, nothing else. Every letter has a one-click unsubscribe link that deletes the record.

← Accept All   Archive
METRResearch

Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT

07.21

We propose a measure of an AI agent’s optimization ability with an “expenditure horizon.” We give an empirical illustration from the NanoGPT speedrun.

Read at METR ↗

More from METR on Accept All

More from METR on Accept All.