INDEX

Explanations

explains fields and their follow-up terms

The neuron is primarily triggered by numeric tokens—especially measurements, statistics, and other numbers (including decimals and large counts).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

are

-1.59

 they

-1.46

 series

-1.41

NESOTA

-1.33

 naszych

-1.31

 pēc

-1.29

 nowy

-1.29

 Obtén

-1.28

却又

-1.27

 naše

-1.24

POSITIVE LOGITS

 amélior

1.30

and

1.20

 причиной

1.06

стью

0.99

 Etats

0.98

yskland

0.98

什么呢

0.98

伊朗

0.97

?».

0.96

!».

0.96

Activations Density 0.253%