INDEX

Explanations

most common classifications

The neuron activates on descriptive frequency or prevalence terms—words that quantify or indicate how common something is (e.g. “common,” “most,” “often,” “frequent,” “examples”).

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

two

-2.48

Both

-2.36

 beiden

-2.28

 both

-2.25

 Both

-2.09

 दोनों

-1.80

両

-1.77

 keduanya

-1.73

Two

-1.72

TWO

-1.70

POSITIVE LOGITS

all

2.30

どれも

2.14

 allemaal

1.76

 всички

1.66

 όλα

1.63

这些

1.60

 wszystkie

1.55

 kaikki

1.52

いずれ

1.52

這些

1.50

Activations Density 0.188%