INDEX

Explanations

mathematical expressions

The neuron activates on numeric tokens (numbers and numerical constants) within mathematical expressions.

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

-1.47

que

-1.45

 from

-1.38

that

-1.37

aka

-1.35

of

-1.34

 که

-1.33

('

-1.31

we

-1.28

typically

-1.27

POSITIVE LOGITS

its

1.81

 її

1.63

 after

1.57

⦊

1.56

álva

1.53

我对

1.52

 Trouvez

1.45

éget

1.45

 Trouver

1.35

 食事

1.35

Activations Density 0.093%