GPT-3 175B (davinci)

3 benchmarks · 4 observations · 3 sources · released 2020-05-28

Scores at a glance

3 benchmarks · large dot: GPT-3 175B (davinci); small dots: closest models where reported

GSM8K ↗9%

GPT-3 175B (davinci) · GSM8K · EM · % ↗

HellaSwag ↗77.5%

GPT-3 175B (davinci) · HellaSwag · Overall accuracy · % ↗

METR task performance & time horizons ↗16.19%

GPT-3 175B (davinci) · METR task performance & time horizons · Average task accuracy ↗

1 more reported result

0.14406 minutes · GPT-3 175B (davinci) · METR task performance & time horizons · 50% success task horizon · METR-Horizon-v1.0 · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

There are not enough shared scored settings to identify close comparisons reliably.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

Holistic Evaluation of Language Models (HELM) ↗
Stanford Center for Research on Foundation Models
Holistic Evaluation of Language Models (HELM) ↗
Stanford Center for Research on Foundation Models
Measuring AI Ability to Complete Long Software Tasks ↗
METR

Names used in source tables

GPT-3 175B (davinci)