INDEX

Explanations

and so on

The neuron fires on the recurring continuation phrase “so on,” i.e. discourse markers indicating “and so on.”

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

and

-1.34

siones

-1.34

jej

-1.26

在其

-1.24

صادر

-1.24

はその

-1.20

 ebenfalls

-1.19

 الرسمي

-1.19

 seen

-1.19

 still

-1.19

POSITIVE LOGITS

,”

1.57

czeń

1.49

 originalen

1.48

 beliebtesten

1.43

]);

1.41

所谓

1.38

،

1.33

}*/

1.32

.”

1.32

จะ

1.31

Activations Density 0.016%