INDEX

Explanations

food preparation cuts

The neuron highlights ingredient‐preparation action words (e.g., peeled, sliced, diced, chopped, cubed, grated).

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

ll

-1.56

 khủng

-1.42

KommentareTeilen

-1.33

레스

-1.27

上的

-1.27

蕯

-1.23

しても

-1.22

on

-1.22

 gaun

-1.22

 broder

-1.20

POSITIVE LOGITS

of

1.59

岨

1.41

逻

1.34

 Länge

1.34

𝓙

1.33

\&

1.31

 cortas

1.30

 atau

1.29

事は

1.28

our

1.27

Activations Density 0.006%