INDEX

Explanations

names followed by actions or possessions

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

and

-3.47

 himself

-2.31

his

-1.88

or

-1.80

-1.43

 خودش

-1.38

he

-1.33

-1.29

 seluruh

-1.26

 ему

-1.25

POSITIVE LOGITS

 sowie

1.93

 respectively

1.92

 Their

1.91

their

1.89

 그리고

1.84

 båda

1.84

Their

1.82

それぞれ

1.81

 both

1.79

are

1.78

Activations Density 0.082%