INDEX

Explanations

logs by

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

 Hampton

-0.08

Pax

-0.08

 stadiums

-0.07

asia

-0.07

 raging

-0.07

eci

-0.07

Kam

-0.07

 Manufact

-0.07

porter

-0.07

 POLITICO

-0.07

POSITIVE LOGITS

 Crab

0.11

ansi

0.11

 incest

0.10

      ↵      ↵

0.10

 Beaver

0.10

 prose

0.10

 mutil

0.10

 storia

0.10

 accident

0.10

isms

0.10

Activations Density 0.001%