INDEX

Explanations

store, deposited, accumulated

The neuron detects floating‐point numbers (numerical values containing decimal points) 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

 melaksanakan

0.84

 실시

0.81

 انجام

0.80

実施

0.80

 gerçekleştir

0.77

 നടത്ത

0.75

 dilakukan

0.75

 gerçekleşt

0.75

实施

0.74

 durchgeführt

0.73

POSITIVE LOGITS

 deposited

0.91

 stored

0.90

 accumulated

0.87

 collected

0.86

 exchanged

0.85

 emitted

0.84

 transported

0.82

 amassed

0.81

stored

0.79

 extracted

0.78

Activations Density 0.378%