INDEX

Explanations

The neuron fires strongly on standalone numeric measurements or decimal‐format numbers embedded 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

No Comments

Negative Logits

 aggiungere

0.79

 install

0.78

 installers

0.76

ใช่

0.73

 renaming

0.73

 instal

0.72

安装

0.72

 isang

0.72

 augmente

0.71

 này

0.71

POSITIVE LOGITS

ه

0.89

емость

0.86

iž

0.84

льные

0.80

рные

0.80

0.79

 প্রাস

0.79

р

0.78

čnih

0.77

льная

0.77

Activations Density 0.000%