INDEX

Explanations

is/are/were followed by descriptor

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

 både

-2.28

 zowel

-2.25

 nejen

-2.08

 sekä

-2.02

不仅

-1.98

either

-1.93

 neither

-1.88

 zarówno

-1.87

 either

-1.86

both

-1.75

POSITIVE LOGITS

 simply

2.61

 just

2.17

 вообще

1.97

 simplemente

1.94

 просто

1.85

 even

1.81

 наоборот

1.73

 other

1.70

 lainnya

1.62

 Perhaps

1.60

Activations Density 0.159%