INDEX

Explanations

born or named

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

odi

-0.09

 Anchor

-0.09

 Imag

-0.09

 untranslated

-0.09

 usable

-0.08

oman

-0.08

 fica

-0.08

 nawet

-0.08

wk

-0.08

 decoded

-0.08

POSITIVE LOGITS

 born

0.16

 Born

0.13

 originally

0.12

 weary

0.12

born

0.11

 bored

0.11

Mu

0.11

 once

0.11

Born

0.11

 Originally

0.10

Activations Density 0.054%