INDEX

Explanations

median and average statistics

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

CONTINUE

-0.94

 WARN

-0.75

SUPPORT

-0.74

 CONTINUE

-0.74

iming

-0.73

тном

-0.70

ddd

-0.69

Continued

-0.69

andescent

-0.68

 śred

-0.68

POSITIVE LOGITS

aronne

0.71

 nắm

0.68

textrm

0.66

 страшно

0.63

뤄

0.63

 evaluation

0.61

红色

0.60

 insurance

0.60

 including

0.60

كثر

0.59

Activations Density 0.065%