INDEX

Explanations

brent

The neuron consistently fires on the token “Brent,” marking mentions of that proper noun.

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

’

-2.63

}\

-2.63

驚き

-2.55

たまた

-2.55

 fervent

-2.48

 celebrid

-2.41

ホント

-2.41

 overarching

-2.39

 myriad

-2.38

罴

-2.38

POSITIVE LOGITS

逦

2.58

獁

2.52

 have

2.36

(),

2.34

while

2.30

婞

2.30

 ragaz

2.27

to

2.23

 tulum

2.23

ственная

2.19

Activations Density 0.002%