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

pla

-0.11

 Voll

-0.09

âĦ

-0.09

 rake

-0.09

 Lamb

-0.09

earch

-0.09

 ancest

-0.08

 Weaver

-0.08

 Thomson

-0.08

Dud

-0.08

POSITIVE LOGITS

bell

0.09

ller

0.09

ÑģÐ¿

0.09

tu

0.09

Vent

0.09

bridge

0.09

olian

0.09

 drip

0.09

 leverage

0.09

/mark

0.09

Activations Density 0.000%

No Known Activations

This feature has no known activations.