INDEX

Explanations

use to describe methods of study

The neuron strongly activates on occurrences of the verb “use,” especially in methodological or purpose statements.

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

还有

-1.58

還有

-1.50

鴬

-1.49

When

-1.43

 églises

-1.42

日時

-1.39

𝓁

-1.39

 ſch

-1.39

</h3>

-1.38

POSITIVE LOGITS

to

2.73

 that

2.28

of

1.76

but

1.50

 this

1.42

 Eğer

1.32

 svoju

1.32

 svou

1.31

 nordrhein

1.30

 Мексика

1.30

Activations Density 0.141%