INDEX

Explanations

string formatting with %

The neuron strongly activates on “%” tokens, especially when they’re used as string‐formatting placeholders.

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

»,

-0.99

²,

-0.97

»،

-0.95

 €,

-0.88

"",

-0.82

',

-0.79

(),

-0.79

€,

-0.78

…,

-0.78

\",\"

-0.78

POSITIVE LOGITS

❩

0.85

-%

0.83

"*

0.76

 ansvar

0.76

RITION

0.75

 pubblici

0.75

 kuten

0.75

дове

0.73

рое

0.73

(_)

0.72

Activations Density 0.003%