INDEX

Explanations

quotes or statements related to allegations and claims in discussions of misconduct

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

 alike

-0.07

 simply

-0.06

 Cous

-0.06

 Uncomment

-0.06

VES

-0.06

announce

-0.06

annonce

-0.06

åĨ

-0.05

ael

-0.05

 Donne

-0.05

POSITIVE LOGITS

 allegedly

0.10

 according

0.09

 alleged

0.08

 laut

0.07

 reportedly

0.07

According

0.07

 court

0.07

according

0.07

 à¤ķà¤¥

0.07

 According

0.07

Activations Density 0.010%