INDEX

Explanations

percentages and comparisons

The main thing this neuron does is detect numeric tokens (counts, percentages, and other multi‐digit figures) in the text.

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

-1.13

itha

-1.11

 enough

-1.11

はこちらから

-1.01

 EXERC

-0.98

をご覧

-0.94

 delgada

-0.93

屋の

-0.92

ええ

-0.91

の準備

-0.91

POSITIVE LOGITS

in

1.72

 earlier

1.30

 previous

1.30

of

1.25

 other

1.16

for

1.12

 others

1.09

legging

1.07

そうで

1.01

 सामान्य

1.00

Activations Density 0.016%