INDEX

Explanations

boolean comparisons or assignments

The neuron activates on literal constants—especially numeric literals (e.g. standalone numbers and numeric tokens).

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

 Autó

-1.52

Mě

-1.37

Ș

-1.34

 biß

-1.33

ྜ

-1.30

 passada

-1.30

 inwest

-1.28

NEYS

-1.27

凩

-1.26

 takie

-1.24

POSITIVE LOGITS

 also

1.55

 también

1.36

 cũng

1.30

 тоже

1.29

 μυ

1.26

并

1.23

خابات

1.20

ellant

1.20

 също

1.19

šče

1.16

Activations Density 0.011%