INDEX

Explanations

references to specific physical attributes or characteristics

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

iled

-0.07

Ã¶g

-0.06

ÑģÐ»Ð¸

-0.06

Ã¡

-0.06

 Cafe

-0.06

oksen

-0.06

igin

-0.06

áº£

-0.06

iedo

-0.05

ASH

-0.05

POSITIVE LOGITS

 disob

0.07

acro

0.07

rippling

0.06

 DISP

0.06

à¤²à¤¤

0.06

 fixture

0.06

Ä±ÅŁÄ±k

0.06

å¹³

0.06

-common

0.06

ç½®

0.06

Activations Density 0.018%