INDEX

Explanations

described conditions following linking verbs

The neuron strongly activates on common auxiliary/copula verbs (e.g. “is,” “are,” “were,” “have”) in running text.

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

and

-1.78

rrbracket

-1.21

 cadeira

-1.20

-1.19

itisation

-1.16

ONAUT

-1.13

oooooo

-1.13

thodon

-1.13

 Funktion

-1.13

 những

-1.12

POSITIVE LOGITS

 that

1.78

 zusätzlichen

1.65

 passenden

1.56

 geprü

1.52

 damaligen

1.49

 estrema

1.46

 needed

1.45

 menjelaskan

1.44

to

1.43

 umfass

1.40

Activations Density 0.243%