INDEX

Explanations

specific numerical or scientific variables and terms associated with mathematical equations or models

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

ukt

-0.07

kt

-0.07

 Leban

-0.07

 Transcript

-0.06

kit

-0.06

drs

-0.06

 nÃ©t

-0.06

@js

-0.06

hdl

-0.06

Fur

-0.06

POSITIVE LOGITS

 film

0.10

 films

0.10

fil

0.08

fim

0.08

fi

0.08

 horror

0.08

ipay

0.07

-d

0.07

 television

0.07

ï

0.07

Activations Density 0.005%