INDEX

Explanations

words related to belief, ownership, and societal roles

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

sert

-0.07

Ð²ÐµÐ´

-0.07

 several

-0.07

Ìī

-0.07

 Several

-0.07

xis

-0.06

ara

-0.06

 another

-0.06

Ð°ÑĢÐ°

-0.06

ÑĥÑĢÐ¸

-0.06

POSITIVE LOGITS

 themselves

0.14

or

0.08

 their

0.07

 whom

0.07

 olursa

0.07

their

0.07

 thems

0.07

{}:

0.06

Activations Density 0.110%