INDEX

Explanations

control state

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 methyl

0.54

 odstr

0.54

 encrypted

0.53

 mindestens

0.53

 vollständ

0.52

 kidnapped

0.51

 człowieka

0.51

 barium

0.50

 enanti

0.50

 antidepress

0.50

POSITIVE LOGITS

types

0.52

RTC

0.49

नक

0.48

 करीत

0.48

utie

0.48

GTEST

0.48

ств

0.47

Utilities

0.47

нили

0.47

 الو

0.47

Activations Density 0.000%

No Known Activations