INDEX

Explanations

competition and comparative actions

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

gentes

-0.84

 bothers

-0.82

۶

-0.82

expect

-0.81

debut

-0.80

 angezeigt

-0.79

 complexity

-0.79

 permitir

-0.77

 enam

-0.77

 locais

-0.77

POSITIVE LOGITS

 between

1.16

who

1.13

inist

1.11

谁能

1.08

 Whose

1.06

Whose

1.05

いかに

1.04

 среди

0.99

 among

0.98

 között

0.94

Activations Density 0.056%