INDEX

Explanations

mathematical formulas and symbols

This neuron detects LaTeX markup commands (i.e. the backslash tokens that introduce math/TeX commands).

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

は

-2.55

が

-2.17

を

-2.13

in

-2.09

значно

-2.08

 hunde

-2.02

вня

-1.89

で

-1.88

 consé

-1.84

も

-1.84

POSITIVE LOGITS

and

2.33

 protože

1.91

 若い

1.88

튿

1.80

 cómodo

1.76

馆

1.76

に追加

1.73

 Fußballspieler

1.73

為に

1.69

 όταν

1.68

Activations Density 0.006%