INDEX

Explanations

lives of

The main thing this neuron does is detect occurrences of the word “lives.”

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

-2.95

</i>

-2.61

 That

-2.36

for

-2.28

In

-2.27

 Taking

-2.23

 ursprünglich

-2.23

凩

-2.23

section

-2.20

 beliebtesten

-2.20

POSITIVE LOGITS

圌

2.97

ments

2.86

 ſou

2.73

屨

2.72

栊

2.72

踽

2.72

of

2.67

 viser

2.63

 diki

2.63

茕

2.61

Activations Density 0.007%