INDEX

Explanations

nuclear power and weapons

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

していました

-1.93

 démocr

-1.91

 chré

-1.87

他們

-1.86

吶

-1.82

Кому

-1.79

īgs

-1.78

ཋ

-1.77

廠

-1.77

 länder

-1.74

POSITIVE LOGITS

of

2.03

selbe

1.89

 were

1.84

 einzigen

1.82

↵

1.74

er

1.70

 technically

1.60

 tomando

1.58

 huéspedes

1.55

↵↵↵↵↵↵↵↵↵↵↵

1.54

Activations Density 0.017%