INDEX

Explanations

copyright notice and license

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

,GL

-0.14

NOI

-0.14

 padd

-0.13

 Voor

-0.12

dop

-0.12

 Whitney

-0.12

 Lucia

-0.12

ISI

-0.12

quee

-0.12

ensively

-0.12

POSITIVE LOGITS

Paw

0.11

 Hawth

0.11

 Raven

0.10

 Salem

0.10

Bav

0.09

 RAID

0.09

 Ravens

0.09

RAF

0.09

 Thorn

0.08

 Caul

0.08

Activations Density 1.468%