INDEX

Explanations

words starting with Rad

The neuron fires on any token containing the letter sequence “rad,” i.e. it detects the substring “rad” (in “Rad,” “rad,” “radon,” etc.).

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

鹚

-2.86

</h3>

-2.77

נות

-2.64

迨

-2.53

 ajuta

-2.44

Contd

-2.42

contd

-2.39

驁

-2.31

vantaged

-2.31

).

-2.31

POSITIVE LOGITS

</b>

3.20

ኧ

2.86

ons

2.84

 acclaim

2.72

᷆

2.67

the

2.63

 pierws

2.56

 américaine

2.52

菒

2.45

2.42

Activations Density 0.007%