INDEX

Explanations

ecosystem

The neuron detects the word "ecosystem" and related contextual terms like "between" in technical, business, and environmental contexts.

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

～

-1.70

-1.69

～、

-1.67

 Wurzeln

-1.66

 gewünschten

-1.63

 regain

-1.58

～～～～

-1.56

 Erwartungen

-1.55

！』

-1.55

ーーーー

-1.54

POSITIVE LOGITS

of

2.80

1.70

 estekak

1.57

いだ

1.57

 épices

1.54

后续

1.52

 japonés

1.52

 ruok

1.51

 portugués

1.51

oform

1.51

Activations Density 0.005%