INDEX

Explanations

instances of the word "even."

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

agan

-0.08

agine

-0.07

elow

-0.06

 anytime

-0.06

conte

-0.06

any

-0.06

 anywhere

-0.06

set

-0.06

rats

-0.06

bye

-0.06

POSITIVE LOGITS

 even

0.08

even

0.07

 MORE

0.07

 though

0.07

-more

0.07

wel

0.07

MORE

0.06

Though

0.06

 EVEN

0.06

Activations Density 0.018%