INDEX

Explanations

prepositions then the first

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

 چهارم

-0.93

之前

-0.93

ροσ

-0.88

 Instead

-0.87

 currentIndex

-0.85

౪

-0.84

 penna

-0.82

itaal

-0.81

 другим

-0.81

人的

-0.81

POSITIVE LOGITS

 first

8.31

 první

6.22

 ersten

5.75

 eerste

5.72

 första

5.66

first

5.47

第一个

5.34

 første

5.31

 pierws

5.28

 első

5.16

Activations Density 0.445%