INDEX

Explanations

broadcasting, broadband, broadening

The neuron fires on occurrences of the string “broad,” i.e. the word “broad” or any token beginning with “broad.”

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

КЕ

-0.84

のお話

-0.84

Diagram

-0.82

↕

-0.81

೯

-0.79

 に関する

-0.76

ジェット

-0.76



-0.75

cesis

-0.75



-0.75

POSITIVE LOGITS

 broad

1.66

broad

1.52

Broad

1.48

casted

1.35

 spectrum

1.35

 broadest

1.27

 BROAD

1.24

 Broad

1.23

spectrum

1.21

 shoulders

1.20

Activations Density 0.020%