INDEX

Explanations

programming keywords for indexing or structure

The neuron is detecting single‐character identifiers (e.g. one‐letter variable or table‐alias names) in code snippets.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

.;

0.60

önt

0.60

.');

0.57

 многи

0.56

?");

0.56

=");

0.55

új

0.54

 destinado

0.54

 거고

0.53

')));

0.53

POSITIVE LOGITS

IN

0.65

it

0.64

来说

0.60

 crowds

0.60

il

0.59

in

0.55

 whose

0.55

 ที่

0.53

 Malaysian

0.53

 String

0.52

Activations Density 0.000%