INDEX

Explanations

common words indicating time or state

The neuron activates on expressions that mark measured spans—especially temporal or quantitative ranges (e.g. “from…to,” “throughout,” numbers with units).

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

in

-1.35

大さじ

-1.10

and

-1.08

年から

-1.06

any

-1.03

 will

-1.01

now

-1.00

venidos

-0.97

已经

-0.96

-0.93

POSITIVE LOGITS

was

1.50

 وكان

1.20

đại

1.18

 réessayer

1.16

my

1.13

 éte

1.12

rildi

1.10

řen

1.09

 service

1.09

ᗩ

1.09

Activations Density 0.009%