INDEX

Explanations

communication verbs

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

<bos>

-1.06

 Meksiku

-0.73

 désolés

-0.61

ⓧ

-0.60

vius

-0.59

 Roskov

-0.59

CodeAttribute

-0.57

 blessé

-0.57

ThroughAttribute

-0.56

Inggris

-0.56

POSITIVE LOGITS

 BoxFit

0.70

be

0.66

\{\\

0.66

 fact

0.54

 argue

0.53

setAccessible

0.52

 MainAxisSize

0.50

ete

0.50

 FACT

0.48

bject

0.48

Activations Density 1.366%