INDEX

Explanations

I feel or understand expressing complex states

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thành

0.71

ร่วม

0.67

นัก

0.66

 основным

0.65

มาชิก

0.64

containing

0.62

ภาย

0.61

 الرب

0.60

 الث

0.60

 United

0.60

POSITIVE LOGITS

 dunno

1.32

 believe

1.31

 знаю

1.27

 guess

1.26

不知道

1.24

 suppose

1.24

 guesses

1.23

 denke

1.22

 honestly

1.19

believe

1.19

Activations Density 1.971%