INDEX

Explanations

references to historical artifacts and their significance

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

apore

-0.08

 Patch

-0.07

Patch

-0.07

 patch

-0.07

egl

-0.07

orca

-0.07

shortcode

-0.07

ÐµÐ½Ð·

-0.07

ìķĶ

-0.07

patch

-0.07

POSITIVE LOGITS

 accompany

0.07

 accompanies

0.07

(strpos

0.06

lest

0.06

oyer

0.06

ast

0.06

QM

0.06

IRD

0.06

aar

0.06

ÙĦÙĦ

0.06

Activations Density 0.006%