INDEX

Explanations

bow, nose, stem, tail, end

The neuron activates on numeric measurement tokens containing decimal points (e.g. “52.5,” “14.312,” “110.5”).

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

lijker

-1.46

-1.44

 Geschä

-1.44

американ

-1.34

 hvorfor

-1.27

 Komponist

-1.27

ような

-1.23

 тях

-1.22

 précédente

-1.20

Have

-1.20

POSITIVE LOGITS

of

2.63

 диван

1.52

》：

1.42

楞

1.41

</h2>

1.39

🕗

1.38

 упаков

1.37

chestra

1.36

 dirig

1.36

astify

1.35

Activations Density 0.018%