INDEX

Explanations

score changes and leads

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 leading

-1.38

 Leading

-1.22

lead

-1.22

leading

-1.21

 lead

-1.16

 trailing

-1.14

lag

-1.09

Leading

-1.08

 leads

-1.04

 trailed

-0.98

POSITIVE LOGITS

ländische

0.77

 först

0.76

 enige

0.75

➕

0.73

мок

0.73

வ்

0.73

 två

0.73

Zen

0.73

("'"

0.73

 BRIEF

0.72

Activations Density 0.013%