INDEX

Explanations

is if, is to, is for

explaining or defining something

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

the

0.48

uma

0.45

一项

0.43

all

0.42

 proviso

0.41

 eine

0.40

 certa

0.39

 those

0.38

THE

0.38

 jika

0.38

POSITIVE LOGITS

through

0.36

 ಬಿಯ

0.34

𝟮

0.33

ваць

0.33

 ಕಂಚ

0.33

스타그램

0.33

ICA

0.33

єш

0.33

зий

0.32

ierrez

0.32

Activations Density 0.039%