INDEX

Explanations

scientific text structure markers

The neuron strongly activates on numeric and quantitative tokens—numbers, measurements, assay values, and similar numeric expressions.

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

olu

-1.45

lllllll

-1.38

()!=

-1.37

何故

-1.34

 kurios

-1.32

 хорошая

-1.30

 that

-1.29

 hatta

-1.28

 potrebbe

-1.25

-1.24

POSITIVE LOGITS

by

1.67

six

1.58

 have

1.53

 seven

1.52

 three

1.47

 шокола

1.42

 několik

1.37

 nine

1.33

two

1.31

ilihan

1.30

Activations Density 0.026%