INDEX

Explanations

cycle

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

 cycle

-0.66

 Cycle

-0.62

cycle

-0.61

Cycle

-0.59

Personensuche

-0.57

 CYCLE

-0.55

 OMITBAD

-0.50

ьаж

-0.48

サイクル

-0.47

ESIS

-0.46

POSITIVE LOGITS

 itſelf

0.79

 themſelves

0.74

Efq

0.74

σθαι

0.73

 purpoſe

0.73

 whoſe

0.73

 myſelf

0.72

 himſelf

0.71

 poffible

0.71

GraphicsUnit

0.69

Activations Density 0.078%