INDEX

Explanations

phrases indicating recommendations or suggestions

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

.scalablytyped

-0.09

Ð¿ÑĢÐ°

-0.08

erken

-0.08

_Tis

-0.08

anta

-0.07

@nate

-0.07

Ø¹Ø±

-0.07

_CLOSED

-0.07

ifen

-0.07

lander

-0.07

POSITIVE LOGITS

 consider

0.09

 consideration

0.09

 Consider

0.08

Consider

0.08

opt

0.08

 check

0.07

go

0.07

look

0.07

 look

0.06

 considered

0.06

Activations Density 0.015%