INDEX

Explanations

compens

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Compens

-1.48

compens

-1.41

 compens

-1.39

 compensate

-1.38

 compensating

-1.38

 compensated

-1.29

 compensation

-1.28

 compensatory

-1.22

 Compensation

-1.15

compensation

-1.09

POSITIVE LOGITS

 larmes

0.59

 varandra

0.57

الإنجليزية

0.56

 démocr

0.56

 ſche

0.54

 utveckling

0.53

 تانيه

0.53

 quæ

0.53

 météo

0.52

 barnen

0.52

Activations Density 0.007%