INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÑĤÑĥÑĢÑĥ

-0.07

ffit

-0.07

blr

-0.06

yÃ©

-0.06

æĹĭ

-0.06

 motivated

-0.06

ä¸ĬèµĽåŃ£

-0.06

ppo

-0.06

wright

-0.06

Regs

-0.06

POSITIVE LOGITS

lis

0.08

.swap

0.08

èī²åĪĹ

0.08

dor

0.07

LOAT

0.07

æľŁçļĦ

0.07

lÃ¥

0.07

pector

0.07

æłĹ

0.07

uenta

0.07

Activations Density 0.000%

No Known Activations

This feature has no known activations.