INDEX

Explanations

by-election

The neuron responds to decimal‐formatted numbers (tokens containing numeric digits plus a decimal point).

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

izieren

-1.67

llllll

-1.57

 mereka

-1.57

strpos

-1.48

 tillegg

-1.47

legd

-1.45

泚

-1.42

шому

-1.41

TacToe

-1.41

phalt

-1.40

POSITIVE LOGITS

of

2.97

by

1.88

to

1.85

 instituições

1.63

 until

1.52

 stupid

1.44

 правительства

1.40

 even

1.38

 became

1.36

 despite

1.36

Activations Density 0.124%