INDEX

Explanations

specified

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

 vecka

-0.61

 spéc

-0.59

 présentes

-0.59

 specified

-0.57

 تانيه

-0.56

 écrits

-0.56

there

-0.55

 fidèles

-0.54

 navires

-0.54

 fidé

-0.54

POSITIVE LOGITS

findpost

0.72

tvguidetime

0.69

LookAnd

0.65

 unknownFields

0.65

########.

0.60

colgroup

0.59

WireFormatLite

0.57

StateException

0.55

 techniques

0.55

 kinds

0.54

Activations Density 0.017%