INDEX

Explanations

checks yes or no

The neuron strongly activates on occurrences of the negation “not.”

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

-2.61

えっ

-1.73

дового

-1.64

퐿

-1.61

Они

-1.56

anément

-1.56

 Что

-1.55

どのような

-1.52

ᐕ

-1.49

autogui

-1.46

POSITIVE LOGITS

1.67

1.45

ко

1.44

<i>

1.38

 особенно

1.33

1.30

 recent

1.30

↵

1.28

1.24

Activations Density 0.004%