INDEX

Explanations

curiosity or checking state

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

 Media

-0.08

 rapidly

-0.08

oco

-0.08

 suddenly

-0.08

ple

-0.07

 Herm

-0.07

 differently

-0.07

ivi

-0.07

Media

-0.07

POSITIVE LOGITS

 curiosity

0.19

 curious

0.19

 intrigued

0.18

Cur

0.18

Cur

0.16

 Intr

0.15

 Checking

0.15

cur

0.15

cur

0.15

 Naturally

0.14

Activations Density 0.139%