INDEX

Explanations

assembly registers and hexadecimal

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 spel

-0.79

 Шо

-0.79

ティスト

-0.77

 гр

-0.77

WTS

-0.76

 Draht

-0.75

ведите

-0.74

uther

-0.73

さな

-0.73

是什么意思

-0.73

POSITIVE LOGITS

 unabhängig

0.80

oline

0.78

ноза

0.77

 offensichtlich

0.75

 solcher

0.75

opponent

0.75

 bürger

0.74

 zombies

0.73

ILog

0.73

cije

0.72

Activations Density 0.002%