INDEX

Explanations

student and academic contexts across languages

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

тап

0.90

такт

0.84

 suatu

0.81

高性能

0.80

ワールド

0.79

 ಸಂಪ

0.79

マスター

0.79

 masterful

0.78

동

0.78

แข็ง

0.76

POSITIVE LOGITS

 afectada

1.69

 aggrieved

1.61

 afectadas

1.56

 afectados

1.56

 víctimas

1.51

 afectado

1.49

 complainants

1.40

 पीड़ित

1.38

 victimes

1.37

 harassed

1.35

Activations Density 0.208%