INDEX

Explanations

sitting, seated, or sat

The neuron detects mentions of sitting or being seated (e.g., “sat,” “sitting,” “seated”) and their associated positional context.

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

of

-1.93

籏

-1.58

 homepage

-1.45

 другие

-1.45

 قبل

-1.45

ことができます

-1.43

 deixando

-1.42

PATRICK

-1.42

 پنجم

-1.39

 ซึ่ง

-1.37

POSITIVE LOGITS

自从

1.73

()],

1.59

 Seat

1.55

 sentado

1.54

现在的

1.48

 prendas

1.48

 restauración

1.45

 seats

1.42

まずは

1.40

話

1.39

Activations Density 0.043%