INDEX

Explanations

fried and pickled foods

The neuron detects words naming food preparation or preservation methods (e.g., pickled, confit, fried, grilled).

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

to

-1.48

犼

-1.45

 メニュー

-1.37

 creat

-1.36

 koala

-1.35

vo

-1.30

 vili

-1.30

 効果

-1.30

zbęd

-1.29

 paella

-1.29

POSITIVE LOGITS

。”

1.75

Another

1.68

Since

1.66

Now

1.59

Also

1.57

Then

1.56

Every

1.55

Many

1.51

Still

1.50

These

1.49

Activations Density 0.026%