INDEX

Explanations

defined or refers to rules/practices

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

à¸Ħà¸²

-0.08

åĬłåĭĴ

-0.08

 dota

-0.08

ritti

-0.07

æ¬¡éĺħè¯»

-0.07

 madr

-0.07

Ð·Ð°Ð´

-0.07

gire

-0.07

ç»§ç»Ńåľ¨

-0.07

Ø§Ø¯ÙĬÙĨ

-0.07

POSITIVE LOGITS

çļĦçī¹çĤ¹æĺ¯

0.08

åĳĹ

0.07

raph

0.07

ault

0.06

çī¹çĤ¹æĺ¯

0.06

è¡¡éĩı

0.06

enu

0.06

 defined

0.06

Defined

0.06

 well

0.06

Activations Density 0.232%