INDEX

Explanations

during a period

The neuron detects temporal‐duration markers—words like “during” (and its equivalents in other languages, e.g. “durante”) that introduce a time span.

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

-2.52



-2.47

 vinculado

-2.34

 affects

-2.27

There

-2.20

 enticing

-2.19

 oftentimes

-2.16

Taking

-2.05

 relacionado

-2.03

 preocupado

-2.03

POSITIVE LOGITS

︲

2.86

’

2.73

“

2.69

for

2.69

⚣

2.42

2.41

 awhile

2.27

);

2.25

呖

2.23

та

2.22

Activations Density 0.003%