INDEX

Explanations

The neuron selectively activates on numeric and statistical expressions (e.g. percentages, P-values, measurement figures), i.e. it finds quantitative data in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Whilst

1.62

出来る

1.59

 Whilst

1.55

が無い

1.55

無い

1.53

हरु

1.43

 dimana

1.43

 सुद्धा

1.42

 বেশী

1.41

 welches

1.40

POSITIVE LOGITS

 nong

1.36

 nonequilibrium

1.21

—

1.21

 toward

1.20

 nont

1.20

さまざまな

1.19

 nonuniform

1.19

 nons

1.18

 nontrivial

1.16

 FIGURE

1.10

Activations Density 0.274%