INDEX

Explanations

while/although followed by a clause

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

åĵªæĢķæĺ¯

-0.09

Ø§ÙħØ±

-0.08

iets

-0.08

 atleast

-0.08

è¦ģä¹Īæĺ¯

-0.08

 suffici

-0.08

 EITHER

-0.08

assel

-0.08

ehm

-0.08

èĩ³å°ĳ

-0.08

POSITIVE LOGITS

 there

0.08

çľĭä¼¼

0.07

 technically

0.07

ä¼ļæľīä¸ĢäºĽ

0.07

ç§įç±»ç¹ģå¤ļ

0.07

æľīäºĽ

0.07

there

0.07

herme

0.07

 theoretically

0.06

åŃĺåľ¨ä¸ĢäºĽ

0.06

Activations Density 0.050%