INDEX

Explanations

know by

The neuron strongly activates on standalone numeric tokens—especially decimal or floating-point numbers.

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

коли

0.79

uses

0.75

раз

0.68

 До

0.67

ધી

0.66

zelf

0.64

 поиска

0.64

ined

0.64

гла

0.64

res

0.64

POSITIVE LOGITS

罠

0.79

LEMN

0.77

^{-}

0.73

 aprovechar

0.73

สาม

0.73

 경우

0.73

 aproveitar

0.72

'\

0.72

Aur

0.72

त्रेयी

0.72

Activations Density 0.001%