INDEX

Explanations

introducing notes or facts

The neuron detects discourse markers that introduce explanatory asides in math writing (e.g. “Note that,” “trivially,” “show that”).

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

-1.42

 Futebol

-1.36

 diſ

-1.36

 zove

-1.34

 Which

-1.33

Ū

-1.31

elines

-1.31

 атмос

-1.30

 futuros

-1.30

 betek

-1.30

POSITIVE LOGITS

</em>

1.92

。」

1.76

 that

1.71

 only

1.50

 mengumumkan

1.47

one

1.45

::*;

1.45

 despite

1.41

1.40

当代

1.39

Activations Density 0.013%