INDEX

Explanations

mathematical expressions involving fractions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

å¼łå»º

-0.08

tÃ©

-0.08

jÃ¡

-0.08

enton

-0.07

ampur

-0.07

ä¼´ä½ł

-0.07

inki

-0.07

_tD

-0.07

ä¾Ŀæ³ķèĩªä¸»

-0.07

etter

-0.07

POSITIVE LOGITS

ä¹Łå°±æĺ¯è¯´

0.07

eps

0.07

 crit

0.06

åĢĴ

0.06

 slightest

0.06

éĵ¶å±±

0.06

agu

0.06

min

0.06

ä¹Łå°±æĺ¯

0.06

_critical

0.06

Activations Density 0.008%