INDEX

Explanations

The neuron detects mentions of numeric “count” or metric fields (e.g. views, visits, downloads, comments, shares) often with their accompanying numbers.

numbers and measurements

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

we

-1.19

selection

-1.18

 will

-1.16

for

-1.07

 интерьер

-1.02

to

-1.02

 required

-1.01

جميع

-1.00

waitKey

-0.99

помним

-0.99

POSITIVE LOGITS

 popularity

1.29

Ār

1.05

 ネイビー

1.05

 människor

1.04

ratore

0.99

 CONSE

0.96

getNumberOf

0.95

 FORCES

0.94

惊讶

0.94

伤心

0.94

Activations Density 0.014%