INDEX

Explanations

This neuron detects mentions of formal quantitative metrics or indexes (e.g. “Human Development Index,” “Ranking,” large numeric scores) in report or title contexts.

countries and rankings

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

abrigo

-1.38

mensa

-1.22

 rozmiar

-1.21

gelang

-1.20

 someone

-1.20

 grossa

-1.19

onymie

-1.17

ㅔ

-1.17

 شركة

-1.17

最終

-1.16

POSITIVE LOGITS

 rankings

1.55

וּ

1.55

וּ

1.49

ּוֹ

1.41

 ratings

1.34

 rating

1.27

ция

1.19

举动

1.17

 ranking

1.15

 речо

1.15

Activations Density 0.027%