INDEX

Explanations

specific noun-verb pairings

consulting for information

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

’

0.66

0.56

WORK

0.49

買い

0.47

DAR

0.47

Pre

0.46

OR

0.46

!:

0.46

PAR

0.46

POSITIVE LOGITS

 Sereth

0.52

 depots

0.50

കാശ

0.49

 abstra

0.46

ämme

0.46

 ويل

0.46

 दिसम्बर

0.46

 よる

0.46

ᠰ

0.45

 صفر

0.45

Activations Density 0.001%