INDEX

Explanations

end of sentence punctuation

The neuron is activated by punctuation marks—especially sentence‐ending periods (and to a lesser extent commas)—rather than by words.

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

 solch

-1.70

 TIPO

-1.66

：“……”

-1.63

 Adicion

-1.52

それは

-1.50

冷蔵庫

-1.49

ほんと

-1.48

 Temos

-1.46

 mitte

-1.45

食べ物

-1.45

POSITIVE LOGITS

use

1.63

 uses

1.53

会の

1.50

 your

1.44

 pomysł

1.42

ἄ

1.39

ksiä

1.39

 also

1.38

ὑ

1.36

ModelSerializer

1.36

Activations Density 0.070%