INDEX

Explanations

references to Jewish identity and related concepts

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

istra

-0.08

ays

-0.07

	fflush

-0.07

ocial

-0.07

 coco

-0.07

imos

-0.07

elin

-0.07

ingles

-0.06

uten

-0.06

iat

-0.06

POSITIVE LOGITS

 Pied

0.06

 Medium

0.06

strap

0.06

ÅŁam

0.06

 Phaser

0.06

bat

0.06

orman

0.06

owitz

0.06

 McInt

0.06

lopedia

0.06

Activations Density 0.006%

No Known Activations