INDEX

Explanations

code structure markers

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

 also

-1.88

自体は

-1.71

 показывает

-1.66

Конечно

-1.65

 aynı

-1.63

 требует

-1.52

瀟

-1.49

縝

-1.48

 решила

-1.48

jut

-1.47

POSITIVE LOGITS

and

2.13

 including

1.96

lichte

1.89

 மற்றும்

1.60

who

1.59

 一人

1.59

BUSINESS

1.57

-------

1.55

libat

1.55

 واتس

1.52

Activations Density 0.007%