INDEX

Explanations

phrases indicating contrasting or clarifying statements

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

ast

-0.06

 chap

-0.06

ahas

-0.06

outu

-0.06

 needless

-0.06

Ast

-0.06

uur

-0.06

umar

-0.06

&C

-0.05

åłĤ

-0.05

POSITIVE LOGITS

marvin

0.07

.Undef

0.07

apgolly

0.07

.scalablytyped

0.07

 _______,

0.07

uitka

0.07

,eg

0.07

/testify

0.07

 traf

0.06

Activations Density 0.015%