INDEX

Explanations

The neuron activates on Russian verb forms (particularly infinitives and imperatives).

technical verbs and actions

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

When

-1.52

-1.48

 Когда

-1.47

 рассказала

-1.38

attendre

-1.35

lanır

-1.34

 alway

-1.32

 มัน

-1.27

 зді

-1.27

Allgeme

-1.25

POSITIVE LOGITS

)|(

1.40

 déput

1.40

точный

1.30

 oraz

1.30

стр

1.29

ActionPerformed

1.27

asures

1.26

镲

1.26

avait

1.24

 avea

1.23

Activations Density 0.010%