INDEX

Explanations

outside a place

The neuron activates on the word “outside,” especially when it introduces a location phrase (e.g., “outside the …”).

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

 Saltar

-1.55

</h2>

-1.53

tedì

-1.48

ığında

-1.47

 behulp

-1.43

阉

-1.38

𓍊

-1.37

ichteten

-1.35

儅

-1.34

iM

-1.34

POSITIVE LOGITS

of

2.31

1.98

 outside

1.81

1.70

1.65

te

1.63

1.59

 زیرا

1.56

ers

1.54

 reorgan

1.52

Activations Density 0.013%