INDEX

Explanations

to + verb, ability, intention

The neuron flags occurrences of the infinitive marker “to” (especially in constructions like “able to,” “want to,” “need to,” etc.).

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

 also

-1.95

 alſo

-1.47

 likewise

-1.42

 todėl

-1.38

 dėl

-1.35

亟

-1.30

 injust

-1.27



-1.26

 bringe

-1.26

 accordingly

-1.25

POSITIVE LOGITS

1.73

يل

1.65

1.52

1.49

1.46

 они

1.45

സി

1.44

be

1.38

 вони

1.36

 Schritte

1.36

Activations Density 0.083%