INDEX

Explanations

possessive own

The neuron detects occurrences of the word “own.”

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

for

-1.52

 głównie

-1.40

：

-1.35

 Çünkü

-1.33

ஂ

-1.32

下了

-1.32

 Other

-1.28

芣

-1.28

ఒ

-1.27

 overeen

-1.27

POSITIVE LOGITS

櫚

1.73

mination

1.60

vetrain

1.58

 decorativos

1.55

درسة

1.52

ズニー

1.51

翼翼

1.49

pergillus

1.47

ଽ

1.46

躋

1.45

Activations Density 0.016%