INDEX

Explanations

rhetorical questions of agreement

The neuron is keyed to the “t” in informal negative contractions (–n’t) used in rhetorical “Who…?” questions.

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

or

-1.74

 cristi

-1.47

as

-1.45

</sub>

-1.36

 would

-1.32

 miten

-1.32

 Conexion

-1.30

جير

-1.30

 without

-1.29

 indicating

-1.29

POSITIVE LOGITS

1.50

1.38

("

1.38

—"

1.37

/"

1.31

</h2>

1.28

 XXXII

1.25

('

1.24

но

1.23

("

1.23

Activations Density 0.009%