INDEX

Explanations

early conversion

The neuron strongly activates on nullable/optional‐type markers (e.g. the “?” in Swift type declarations or “OrNil” in Objective-C method names).

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

-1.44

 zehn

-1.43

 seguramente

-1.39

 kuitenkin

-1.38

们

-1.36

ſſer

-1.34

 probablemente

-1.31

es

-1.30

 sugiere

-1.29

 męska

-1.27

POSITIVE LOGITS

寕

1.37

 innych

1.34

 avoit

1.29

錞

1.24

をも

1.23

횃

1.23

vla

1.22

Kirche

1.21

場は

1.20

데

1.18

Activations Density 0.001%