INDEX

Explanations

references to comic book characters and creators

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ikut

-0.08

 Kobe

-0.08

haf

-0.07

legate

-0.07

 Ahmet

-0.07

Wat

-0.07

rada

-0.07

conde

-0.06

elier

-0.06

PIO

-0.06

POSITIVE LOGITS

Hom

0.13

Hom

0.10

 Amazon

0.10

 Boys

0.09

 superhero

0.09

hom

0.09

Sup

0.09

 superheroes

0.09

 amazon

0.09

sup

0.09

Activations Density 0.007%