INDEX

Explanations

Technical content

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

 lèvres

-0.56

 vectoriales

-0.52

">—

-0.51

setViewportView

-0.51

 zelve

-0.49

rohr

-0.48

haviour

-0.47

actéristique

-0.46

 eorum

-0.45

 moeite

-0.44

POSITIVE LOGITS

UserScript

0.61

ніципалі

0.60

cm

0.57

 Monopoly

0.56

 مشين

0.54

 كومونز

0.53

Ang

0.53

例句

0.53

 يتيمه

0.52

pyplot

0.52

Activations Density 0.000%