INDEX

Explanations

elements related to communication and operation indicators

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

oco

-0.06

uggy

-0.06

 Mezi

-0.06

Tes

-0.06

ä¾¡

-0.06

Č↵

-0.06

Å¡ÃŃ

-0.06

elic

-0.06

Zug

-0.06

POSITIVE LOGITS

.Safe

0.08

 impending

0.07

ebra

0.06

æĦŁ

0.06

åĳ

0.06

æĸ¹åĲĳ

0.06

ilib

0.06

 direction

0.06

 status

0.06

 imminent

0.06

Activations Density 0.041%