INDEX

Explanations

existence/location

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 everyone

-1.05

everyone

-0.98

 everybody

-0.87

everybody

-0.84

Everyone

-0.76

 nobody

-0.73

 Everyone

-0.73

 Everybody

-0.70

Everybody

-0.70

Nobody

-0.66

POSITIVE LOGITS

 isComment

0.66

bewerken

0.65

 Else

0.62

 selectivity

0.60

NavController

0.59

isContained

0.57

aneous

0.57

mentable

0.57

наче

0.57

 himo

0.56

Activations Density 0.039%