INDEX

Explanations

removing seeds, bones, and shells

The neuron activates on verbs that describe food-preparation or cooking steps (e.g. peeled, seeded, cut, scraped, discard).

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

豋

-1.57

 Generale

-1.44

ecznie

-1.38

 luxem

-1.34

 publice

-1.31

 Departement

-1.31

 چنین

-1.27

 Escolar

-1.27

 aquarelle

-1.25

 brune

-1.25

POSITIVE LOGITS

ᥬ

1.48

chaleco

1.45

 soutě

1.42

 Павел

1.38

россий

1.38

 następnie

1.37

 obyvateľov

1.37

了过来

1.33

 Then

1.33

 blancas

1.32

Activations Density 0.023%