INDEX

Explanations

family followed by a noun

The neuron is specifically detecting occurrences of the word “Family” (typically as part of organization or section titles).

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>

-2.83

-2.70

</h3>

-2.22

-2.14

</sub>

-2.09

-1.95

-1.93

 passende

-1.87

</u>

-1.71

 technisch

-1.67

POSITIVE LOGITS

潆

2.17

谞

2.13

باره

2.11

 ataupun

2.05

hörigen

1.98

ୌ

1.97

辚

1.96

ﾝ

1.95

üyük

1.91

地说道

1.90

Activations Density 0.009%