INDEX

Explanations

specific terms and symbols related to academic or technical content, particularly in languages other than English

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

ÂĹ

-0.06

 labour

-0.06

rig

-0.05

man

-0.05

---

-0.05

 Laboratory

-0.05

 perd

-0.05

 Techn

-0.05

 laboratory

-0.05

arden

-0.05

POSITIVE LOGITS

 bullying

0.12

bul

0.10

Bul

0.10

 bullied

0.10

 Islam

0.10

CA

0.10

 district

0.10

 bully

0.09

 Muslim

0.09

Islam

0.09

Activations Density 0.002%