INDEX

Explanations

why questions and exclamations

The neuron consistently fires on the very first content word (often a pronoun, noun, or modal) that follows a “Why” question word.

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

-1.68

his

-1.62

 seinen

-1.56

-1.50

ers

-1.45

 diferentes

-1.45

 andere

-1.41

-1.40

-1.38

POSITIVE LOGITS

我们可以

1.68

ены

1.48

腳

1.46

жели

1.46

remadura

1.43

پاسخ

1.39

Inconvénients

1.36

鸢

1.36

="$

1.35

 quitter

1.34

Activations Density 0.007%