INDEX

Explanations

list type, goal, or question

The neuron strongly activates on words that serve as document field labels or section headings (e.g., “type,” “goal”).

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

édric

-1.08

▿

-0.89

 dolu

-0.89

агностика

-0.89

ların

-0.88

fillRect

-0.88

}.\

-0.88

 cancelación

-0.86

bibliography

-0.86

🧳

-0.85

POSITIVE LOGITS

 vitale

1.15

of

1.15

OBJETIVO

1.12

apprendre

1.05

 αυτή

1.02

 poë

1.01

 humeur

1.00

and

0.97

 салат

0.95

歲

0.94

Activations Density 0.000%