INDEX

Explanations

conjunctions and conditional phrases

The neuron fires on JSON-style dependency declarations and version strings (e.g. the “dependencies” key, quoted package names, version numbers, and registry URLs).

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

 this

-1.95

it

-1.87

ったのは

-1.65

 cómoda

-1.63

 there

-1.60

 tohoto

-1.59

 suministro

-1.54

ってください

-1.53

bảng

-1.53

 minutes

-1.52

POSITIVE LOGITS

but

1.71

Whether

1.59

However

1.59

 लेकिन

1.58

If

1.53

Если

1.52

Though

1.52

Although

1.48

 แต่

1.48

 اگر

1.46

Activations Density 0.020%