INDEX

Explanations

via

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

ollow

-0.12

 Williamson

-0.11

unga

-0.10

zman

-0.10

 Scho

-0.10

Kro

-0.10

SVC

-0.10

ZH

-0.10

itty

-0.10

woods

-0.10

POSITIVE LOGITS

 differently

0.20

via

0.16

 separately

0.15

 elsewhere

0.14

 indoors

0.13

 outdoors

0.12

via

0.12

 through

0.11

 ÑĩÐµÑĢ

0.11

Via

0.11

Activations Density 1.071%