INDEX

Explanations

abstract forces at work

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

命

-0.74

imarães

-0.72

inence

-0.70

形式

-0.70

azaki

-0.68

 hilo

-0.68

 Mendelssohn

-0.68

लग

-0.68

gb

-0.68

լ

-0.67

POSITIVE LOGITS

 working

2.31

at

2.22

 work

1.63

ทำงาน

1.60

 trabajando

1.53

 operating

1.52

 hard

1.49

働

1.38

working

1.36

 workings

1.35

Activations Density 0.047%