INDEX

Explanations

enforcement of rules

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

spot

-0.22

tom

-0.21

æĥ³æ³ķ

-0.21

æİ¥è§¦åĪ°

-0.21

spark

-0.21

 inspired

-0.20

VO

-0.20

jected

-0.20

bur

-0.20

çĶ¨æĪ·æıĲä¾Ľ

-0.20

POSITIVE LOGITS

åĬĽåº¦

0.29

 against

0.27

æĢ§å¼º

0.26

 Enforcement

0.25

against

0.25

æĥ©

0.25

 Against

0.24

Against

0.23

/rec

0.23

 stricter

0.23

Activations Density 0.007%

No Known Activations