INDEX

Explanations

varies

The neuron specifically detects occurrences of the word “varies.”

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

be

-2.33

ᕕ

-2.23

羈

-2.23

ra

-2.22

 полиции

-2.16

୮

-2.11

 criterion

-2.08

 ripple

-2.06

eſ

-2.06

퐼

-2.05

POSITIVE LOGITS

vantaged

2.58

独特

2.28

넛

2.23

のデザイン

2.19

')

2.19

臵

2.16

of

2.14

 sozinho

2.13

灩

2.13

 berühm

2.11

Activations Density 0.003%