INDEX

Explanations

or other options

The neuron fires on list-style coordinating conjunctions—especially the word “or” and related punctuation—used to enumerate items (as in legal or formal lists).

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

横に

-1.94

 моей

-1.93

眼前的

-1.77

 hlavně

-1.77

 беше

-1.73

();

-1.70

 jeudi

-1.65

 noires

-1.63

黒い

-1.63

 живет

-1.63

POSITIVE LOGITS

 with

2.84

 other

2.83

or

2.77

on

2.61

 that

2.48

These

2.20

in

2.08

Other

2.06

 cuaca

2.03

其他

2.02

Activations Density 0.165%