INDEX

Explanations

code definitions and configurations

The neuron strongly activates on ALL-CAPS identifier tokens (e.g. preprocessor constants and acronyms), flagging uppercase acronyms or constant 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

cyklo

-1.38

 distritos

-1.34

kug

-1.30

ortheast

-1.30

 caminh

-1.29

 Selecciona

-1.23

 Confira

-1.21

 Trabal

-1.21

mousse

-1.20

 DELLE

-1.20

POSITIVE LOGITS

1.34

te

1.31

()

1.27

at

1.27

ss

1.27

1.25

1.24

1.23

or

1.21

Activations Density 0.010%