INDEX

Explanations

the word "and" appearing in lists, enumerations, or technical contexts.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æľīæĽ´å¤ļçļĦ

-0.09

æĪĳä»¬è¿ĺæĺ¯

-0.08

æĽ´åĬłçļĦ

-0.08

Ø§ÙħØ±

-0.08

æľīæĽ´å¤ļ

-0.08

åĩĮå¤©

-0.08

æĺ¾å¾ĹæĽ´åĬł

-0.08

è¿Ļæł·çļĦæĥħåĨµ

-0.08

æĽ´åĬłæĸ¹ä¾¿

-0.08

æĽ´åĬł

-0.07

POSITIVE LOGITS

 logically

0.07

 evidently

0.06

¯

0.06

 wyra

0.06

åħ¸åŀĭçļĦ

0.06

æł¸å¿ĥ

0.06

æľ¬é¢ĺèĢĥæŁ¥

0.06

 concise

0.06

æľĢåħ³éĶ®

0.06

 clearly

0.06

Activations Density 0.000%

No Known Activations