INDEX

Explanations

mathematical expressions and sequences

The neuron fires on individual numerals (digits) when they appear as part of mathematical sequences or numeric 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

 bedeckt

-1.51

 umgeben

-1.50

 that

-1.49

 prevé

-1.48

-1.46

 motto

-1.45

イヤレス

-1.41

-1.38

 tomado

-1.37

های

-1.36

POSITIVE LOGITS

，

1.60

1.47

1.45

 рассказыва

1.44

 jestli

1.44

ਰੀ

1.43

你

1.41

usstsein

1.41

 говори

1.40

osoba

1.39

Activations Density 0.097%