INDEX

Explanations

praising a job done

The neuron fires strongly on praise phrases about someone’s performance—especially variants of “did a good/amazing/outstanding job.”

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

 verschillen

-1.48

and

-1.37

-1.33

之为

-1.33

 kleuren

-1.29

 inmueble

-1.28

 equipaje

-1.28

 rebajas

-1.28

 tijden

-1.27

فحة

-1.25

POSITIVE LOGITS

as

1.59

of

1.41

mo

1.27

1.19

ic

1.15

gre

1.13

водить

1.13

1.12

Activations Density 0.008%