INDEX

Explanations

issue

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

 pleaſure

-1.10

 myſelf

-1.10

 Jefus

-1.04

 Chriftian

-1.01

 purpoſe

-0.98

 reaſon

-0.96

 ſtate

-0.94

Efq

-0.94

Eſ

-0.94

 ſmall

-0.92

POSITIVE LOGITS

HÀ

0.43

DESCRIPTION

0.42

mT

0.42

RA

0.41

 चुनें

0.41

and

0.41

OSHA

0.41

abetta

0.40

 codice

0.40

omitempty

0.40

Activations Density 0.069%