INDEX

Explanations

these, respective, or certain items

The neuron detects numeric expressions and measurements (e.g. counts, units, percentages, and other figures) throughout 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

 especially

-1.72

to

-1.68

 particularly

-1.52

by

-1.51

and

-1.51

 that

-1.49

 even

-1.38

or

-1.27

不

-1.26

 vicini

-1.20

POSITIVE LOGITS

这些

1.63

 respectively

1.46

 jeweils

1.45

それぞれ

1.45

這些

1.41

these

1.33

いずれ

1.29

Эти

1.29

 acestea

1.26

这两

1.24

Activations Density 0.146%