INDEX

Explanations

robbery and threats

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åıįçľģ

-0.27

ä¹Ŀå¤§

-0.27

åŃķèĤ²

-0.27

è¿ĻæĿ¡è·¯

-0.25

 Parenthood

-0.24

æ¯ıä¸Ģæ¬¡

-0.23

-this

-0.23

 nextState

-0.23

ä»£è¡¨çĿĢ

-0.22

èĪªæµ·

-0.22

POSITIVE LOGITS

è´¢çī©

0.33

çİ°éĩĳ

0.30

 robbery

0.28

æĬ¢åĬ«

0.28

 robber

0.26

åĪ°è´¦

0.26

å½ĵåľº

0.25

æĮ£

0.24

çĽĹçªĥ

0.24

åºĹåĨħ

0.24

Activations Density 0.039%