INDEX

Explanations

by Name

The neuron strongly activates on floating‐point numeric tokens (numbers with decimal points).

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

and

-1.63

 yaptığı

-1.55

.$

-1.31

※

-1.27

hoc

-1.25

🇹

-1.25

 часть

-1.23

CommonModule

-1.23

issippi

-1.22

ığı

-1.22

POSITIVE LOGITS

dellín

1.46

鱻

1.46

way

1.45

</em>

1.40

 occasions

1.39

tismo

1.39

 پژ

1.38

いいね

1.38

 ダーク

1.36

 Ajoutez

1.34

Activations Density 0.024%