INDEX

Explanations

API access and promises

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

 Modify

-0.67

Modify

-0.66

 Store

-0.66

ói

-0.65

Ⓥ

-0.63

少女

-0.63

Store

-0.62

ochemical

-0.62

 Keep

-0.61

Invalidate

-0.61

POSITIVE LOGITS

Auf

0.75

ABCDEF

0.75

Retrieval

0.73

一堆

0.71

flam

0.70

 Siegel

0.70

gos

0.70

 Profil

0.70

 assemb

0.69

Profil

0.69

Activations Density 0.059%