INDEX

Explanations

questions and conjunctions

The neuron activates on numeric tokens (integers or decimals) in the text.

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

ungkapkan

-1.43

篋

-1.39

 they

-1.38

🅻

-1.38

 irgend

-1.38

 rainha

-1.35

 discípulos

-1.34

有何

-1.34

 měla

-1.33

 apoi

-1.32

POSITIVE LOGITS

to

1.99

and

1.64

the

1.55

mathrm

1.54

If

1.44

setupUi

1.37

 Guevara

1.37

*{-

1.35

â

1.34

or

1.34

Activations Density 0.071%