INDEX

Explanations

Research papers

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UserScript

-0.73

NOPQRST

-0.68

AndEndTag

-0.68

PerformLayout

-0.67

 citoy

-0.59

ClientSize

-0.59

OrNil

-0.59

Viki

-0.58

awtextra

-0.57

 edelstahl

-0.57

POSITIVE LOGITS

0.45

qu

0.40

la

0.40

getL

0.40

pmd

0.39

lent

0.39

lcc

0.38

laq

0.38

lb

0.37

llum

0.37

Activations Density 0.000%