INDEX

Explanations

either/or constructions

The neuron spotlights standalone numeric tokens (numbers) in the text.

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

and

-1.85

corrhi

-1.81

vejec

-1.62

konos

-1.54

ようになりました

-1.51

fatter

-1.49

 måneder

-1.48

archiwizowane

-1.44

鋮

-1.41

Dabei

-1.40

POSITIVE LOGITS

 they

1.66

は

1.52

を

1.49

1.34

と

1.32

も

1.31

が

1.30

醍

1.25

1.24

1.22

Activations Density 0.030%