INDEX

Explanations

number of times

The neuron fires on words and phrases that express counts or frequencies (e.g. “hundreds,” “thousands,” “a few,” “a dozen,” “multiple times”).

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

 that

-2.14

 from

-2.02

 which

-1.80

 such

-1.72

-1.64

and

-1.44

huo

-1.38

 having

-1.32

 their

-1.31

from

-1.30

POSITIVE LOGITS



1.58

艽

1.56

煳

1.55

 ögon

1.52

}],

1.50

కర్య

1.49



1.48

皲

1.48

鹉

1.47

 merupakan

1.46

Activations Density 0.057%