INDEX

Explanations

Amazon Senate Steph net Concern

The neuron detects numerical tokens (plain numbers or numeric expressions) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

</strong>

-2.25

 hoher

-1.92

 hohem

-1.88

 this

-1.78

 geringen

-1.71

理由は

-1.65

 arquitecto

-1.64

 větší

-1.59

 unangemess

-1.59

if

-1.55

POSITIVE LOGITS

 mereka

1.94

and

1.82

et

1.72

 setiap

1.62

 Straße

1.50

{~

1.47

𝑥

1.46

llten

1.43

🌎

1.42

années

1.41

Activations Density 0.018%