INDEX

Explanations

once again

The neuron strongly activates on the repeated transitional phrase “once again.”

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

usitis

-1.42

 навіть

-1.42

⾼

-1.40

 After

-1.33

 Both

-1.30

 llevan

-1.28

Isn

-1.27

 після

-1.26

歿

-1.26

 after

-1.25

POSITIVE LOGITS

they

1.38

צ

1.38

 временем

1.37

他的

1.37

തി

1.34

ଢ

1.33

ڑ

1.32

durch

1.30

кер

1.26

rà

1.25

Activations Density 0.007%