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

ABCDE

-0.10

arte

-0.08

kee

-0.08

kor

-0.08

antino

-0.08

ped

-0.08

AGO

-0.08

 wiring

-0.08

 resist

-0.08

 Bravo

-0.08

POSITIVE LOGITS

 Bere

0.09

btw

0.09

 Cecil

0.09

OT

0.09

ergy

0.09

ipur

0.09

ãĥī

0.09

 Dale

0.08

 delicate

0.08

abeth

0.08

Activations Density 0.000%

No Known Activations

This feature has no known activations.