INDEX

Explanations

fragmented information

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

 most

-1.15

most

-0.92

Most

-0.75

 MOST

-0.69

 Most

-0.67

MOST

-0.66

unk

-0.54

 EventEmitter

-0.53

Становништво

-0.52

ณ์

-0.51

POSITIVE LOGITS

sizeCache

0.70

:✨

0.69

 métallique

0.66

findpost

0.61

^(@)

0.60

 paroisse

0.60

 ſhe

0.59

Efq

0.59

illage

0.59

RuleContext

0.59

Activations Density 1.629%