INDEX

Explanations

examine areas

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.49

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.47

Club

0.46

Procedure

0.45

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.45

Arrow

0.44

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.44

Munic

0.44

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.43

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.43

POSITIVE LOGITS

 hepatocellular

0.61

Heb

0.52

 Cato

0.52

ateful

0.51

 hues

0.51

emonic

0.51

 धूम

0.50

 जिन्हें

0.49

ated

0.48

igkeit

0.48

Activations Density 0.000%