INDEX

Explanations

gift for

The main thing this neuron does is detect mentions of “gift” (and “gifts”) in promotional or product-description text.

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

творення

-1.45

Kelebihan

-1.41

 Prefeitura

-1.39

Ⴈ

-1.38

aming

-1.35

Ւ

-1.34

Добрый

-1.34

</strong>

-1.32

Приготовление

-1.32

сті

-1.30

POSITIVE LOGITS

 that

1.74

or

1.61

for

1.41

of

1.38

dpy

1.36

was

1.34

 gezondheid

1.30

 alanı

1.28

 maravillosas

1.27

In

1.27

Activations Density 0.016%