INDEX

Explanations

numbers in the millions

The neuron selectively responds to numeric tokens representing measured values—especially decimals and percentages.

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

 Agregar

-1.52

 какъ

-1.47

 esquí

-1.41

 作る

-1.39

ího

-1.38

Ако

-1.37

各样的

-1.36

tied

-1.35

 úspě

-1.34

鶲

-1.34

POSITIVE LOGITS

1.62

 million

1.43

 their

1.39

 years

1.37

 flera

1.37

？』

1.32

 from

1.31

">

1.31

だった

1.26

1.24

Activations Density 0.090%