INDEX

Explanations

mathematical groups and classes

This neuron detects exponentiation notation, i.e. caret symbols (^) followed by superscript numbers.

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

 then

-1.77

not

-1.68

</b>

-1.62

 must

-1.61

To

-1.56

\&

-1.53

-1.50

not

-1.48

);

-1.45

POSITIVE LOGITS

Kelas

1.76

 Sociale

1.61

Gruppe

1.61

ungguhnya

1.60

istische

1.60

㍑

1.59

 vín

1.58

 ſte

1.57

𝖾

1.56

Dá

1.55

Activations Density 0.059%