INDEX

Explanations

concepts followed by prepositions

The main thing this neuron does is detect numeric expressions (quantities, measurements, percentages, and other numbers).

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

 Fähigkeit

-1.76

bietet

-1.59

我們

-1.59

 Erfüllung

-1.59

푠

-1.56

ﺶ

-1.56

 rafraî

-1.55

私たち

-1.55

淏

-1.52

Aktual

-1.52

POSITIVE LOGITS

of

2.92

for

2.48

 with

1.98

at

1.97

as

1.91

to

1.80

↵↵↵↵

1.66

↵↵↵↵↵

1.59

 also

1.58

by

1.55

Activations Density 0.161%