INDEX

Explanations

The neuron is primarily triggered by numeric tokens (individual digits or multi‐digit numbers).

numbers and bases

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

 bomberos

-1.10

が違う

-1.09

 somente

-1.05

でしょうね

-1.04

もとも

-1.02

possibly

-0.99

がありますが

-0.98

菒

-0.98

 ugyan

-0.97

却没有

-0.97

POSITIVE LOGITS

1.60

1.47

1.45

1.30

to

1.27

 when

1.27

 with

1.23

1.22

1.15

1.14

Activations Density 0.175%