INDEX

Explanations

masters and teachers

This neuron activates on words naming pedagogical roles or relationships—terms like “teacher,” “student,” “pupil,” or similar instructional titles.

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

 meinem

-1.34

 recklessly

-1.33

-1.30

-1.29

慙

-1.23

-1.20

decir

-1.17

Lun

-1.16

-1.15

 أخرى

-1.15

POSITIVE LOGITS

🅃

1.62

不

1.42

☂

1.39

ｔ

1.38

 понадоби

1.38

 Hidup

1.36

 podia

1.36

🅣

1.34

ادن

1.34

 that

1.33

Activations Density 0.031%