INDEX

Explanations

black

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

 Houſe

-0.59

 becauſe

-0.57

 ſeveral

-0.57

 ſta

-0.57

 ſmall

-0.56

Efq

-0.56

 occaf

-0.56

 nahilalakip

-0.56

 iſt

-0.54

$=-

-0.54

POSITIVE LOGITS

<bos>

0.72

 مشارکت‌کنندگان

0.51

setHorizontal

0.44

 női

0.43

onAttach

0.43

 dönt

0.42

dom

0.42

top

0.42

doms

0.42

YNC

0.42

Activations Density 0.010%