INDEX

Explanations

each time, every time, multiple times

The neuron detects discourse‐marker words—small function words (like “this,” “again,” “but,” “every,” “time,” etc.) that signal temporal or contrastive transitions.

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

зон

0.52

ॢ

0.51

esque

0.50

两大

0.48

 każdego

0.48

全年

0.48

quet

0.47

IMO

0.47

ovaných

0.46

زمین

0.46

POSITIVE LOGITS

 times

2.75

 fois

2.52

 veces

2.47

 time

2.44

 vezes

2.30

 lần

2.27

times

2.22

ครั้ง

2.13

time

2.02

vez

1.95

Activations Density 0.429%