INDEX

Explanations

differentiability classes C^n

This neuron activates on tokens belonging to mathematical/LaTeX notation (math delimiters, symbols, numbers, and formula elements).

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

forbes

-0.79

ūnas

-0.78

Ory

-0.75

ory

-0.74

aser

-0.72

解除

-0.71

 Rezept

-0.70

hrer

-0.70

scl

-0.70

 arom

-0.69

POSITIVE LOGITS

 smoothness

1.09

 regularity

1.01

 class

0.91

 Smooth

0.81

 oignons

0.80

 générale

0.78

 adik

0.78

Twice

0.78

Smooth

0.77

 klasse

0.77

Activations Density 0.026%