INDEX

Explanations

software and app development

The neuron fires on code‐specific tokens—especially filenames, file extensions, and configuration keys (e.g. “App.css,” “index.html,” manifest fields, “Create React App,” etc.).

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

ときは

-1.03

 cadeaux

-1.02

IntentFilter

-1.02

tA

-1.02

 microfibra

-1.02

 katalog

-1.02

 rappresenta

-1.01

isman

-1.01

Wyd

-1.01

日起

-0.99

POSITIVE LOGITS

Like

1.13

Many

1.11

Now

1.07

Despite

1.01

at

0.99

Nearly

0.99

Although

0.97

ｯ

0.96

GRAFÍA

0.96

Both

0.93

Activations Density 0.000%