INDEX

Explanations

multiplication

The neuron fires on numeric tokens—especially floating‐point or decimal measurements—i.e. it spots numbers (values with digits, decimals, units, etc.) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 वित्तीय

0.86

 उत्सव

0.83

聩

0.81

 Siamo

0.80



0.77

 financiero

0.76

 नैतिक

0.74

 Comercio

0.73

 বিদ্ব

0.73

一篇

0.73

POSITIVE LOGITS

×

1.64

×

1.59

1.36

1.32

 times

1.30

乘以

1.28

 multiplied

1.26

1.20

times

1.18

*(

1.16

Activations Density 0.098%