INDEX

Explanations

the word "help" and words commonly associated with it

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

Assistance

-1.10

assistance

-1.08

 Assistance

-1.07

 assistance

-1.02

 ASSISTANCE

-0.91

 Assist

-0.90

Assist

-0.88

assist

-0.86

ASSIST

-0.86

 assisting

-0.85

POSITIVE LOGITS

olesterol

0.57

 interest

0.56

 thema

0.54

Tarih

0.54

 beho

0.54

 interests

0.53

HashCode

0.53

Blo

0.52

PostConstruct

0.51

 profile

0.50

Activations Density 0.020%