INDEX

Explanations

list separators

The neuron strongly activates on the word “as” when it introduces a clause of the form “we as …” or “we, as …,” i.e. subordinating-conjunction uses of “as” in identity statements.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

().

0.47

jenigen

0.46

).

0.45

».

0.43

 trasform

0.43

ության

0.42

 வகையில்

0.42

 transform

0.42

డంతో

0.42

POSITIVE LOGITS

न्ही

0.55

 پورا

0.54

뎁

0.52

ენი

0.51

 cały

0.50

騇

0.50

álně

0.50

zyła

0.49

އ

0.49

 घेतला

0.48

Activations Density 0.036%