INDEX

Explanations

discussing specific entities or items

the neuron activates on semantically rich content words—mostly multi-syllabic nouns and adjectives that carry the main topics (e.g. “problems,” “amazing,” “billions,” “bands”).

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

しかも

-1.14

约定

-1.02

 замена

-1.00

幹線

-1.00

 literaria

-0.99

sp

-0.98

 sacerdotes

-0.96

 直接

-0.94

 cotidiana

-0.93

POSITIVE LOGITS

and

1.47

際は

1.17

際には

1.10

 punggung

1.09

According

1.09

！”

1.05

感じで

1.03

/*!

1.03

ivasan

1.00

Leírás

1.00

Activations Density 0.086%