INDEX

Explanations

app features and software

The neuron detects mentions of specific app interface elements or feature sections (e.g. vocabulary section, leaderboard, exercise types).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

but

-1.35

而非

-1.16

 computadoras

-1.15

 zamiast

-1.13

 rather

-1.13

トム

-1.11

 gimnasio

-1.11

 instead

-1.10

 plutôt

-1.10

}-

-1.09

POSITIVE LOGITS

 COMPR

1.35

✸

1.32

 your

1.31

 RECOMEND

1.27

 compagnies

1.25

 milliard

1.25

 TECHNI

1.23

 habang

1.22

ଶ

1.22

 Prefeitura

1.20

Activations Density 0.340%