INDEX

Explanations

No Explanations Found

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

go

-0.11

leur

-0.10

Ø§Ø³

-0.10

dig

-0.09

Ã¶rt

-0.09

IMS

-0.09

 routinely

-0.09

ull

-0.09

lore

-0.09

 hopeless

-0.09

POSITIVE LOGITS

-only

0.09

 outnumber

0.09

oud

0.09

Bulk

0.09

sne

0.09

 uniqu

0.09

 desta

0.09

 mere

0.09

 sneak

0.09

tro

0.08

Activations Density 0.000%

No Known Activations

This feature has no known activations.