INDEX

Explanations

location attributes and values

The neuron is looking for location‐ or position‐related terms (e.g. “LOCATION,” “located,” coordinate attributes like “href” URLs pointing to a resource).

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

找不到

-1.14

 Gouver

-1.13

 botão

-1.07

でに

-1.05

लिए

-1.02

 fortale

-1.02

⃨

-1.00

 এবং

-0.98

 জন্য

-0.98

 手工

-0.98

POSITIVE LOGITS

('');

1.09

خواه

1.01

layın

1.00

ेमाल

0.98

achieve

0.97

0.93

in

0.92

bó

0.91

 LIABLE

0.91

这位

0.90

Activations Density 0.000%