INDEX

Explanations

references to denial or dismissals of claims and allegations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

 salope

-0.07

æı

-0.07

mbH

-0.07

acific

-0.07

ToSelector

-0.07

LocalizedString

-0.06

 ÑĥÐ¼Ð¾Ð²

-0.06

osloven

-0.06

Ãºde

-0.06

blick

-0.06

POSITIVE LOGITS

 being

0.11

 having

0.10

 ever

0.09

any

0.08

being

0.07

 that

0.07

isex

0.07

 Being

0.06

 claims

0.06

Jun

0.06

Activations Density 0.004%

No Known Activations