INDEX

Explanations

TORTIOUS BEHAVIOR

facility location, classification, or error detection

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

Nar

0.43

Hux

0.42

yser

0.41

Nar

0.41

to

0.40

VX

0.40

 Vict

0.40

 गठबंधन

0.40

 লো

0.39

 Witt

0.39

POSITIVE LOGITS

 wrong

0.53

 универса

0.49

 liberty

0.45

 moch

0.45

 mauvaise

0.44

 universality

0.44

뚤

0.44

 Wrong

0.44

 libertà

0.44

Wrong

0.43

Activations Density 0.000%