INDEX

Explanations

actions after certain words

failure to prevent

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

OK

0.53

DK

0.49

Cnt

0.48

 brisk

0.47

activity

0.47

 warranted

0.47

ка

0.46

ac

0.46

TOF

0.46

ent

0.45

POSITIVE LOGITS

 Hölder

0.53

 ConcurrentTask

0.52

 Veranst

0.50

 Constantin

0.49

 Tijdens

0.48

 Gül

0.48

 製造

0.48

 Introductory

0.48

 Determining

0.47

 スペ

0.47

Activations Density 0.000%