INDEX

Explanations

circular motion or rotation

The main thing this neuron does is find words describing rotational or revolving motion.

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

 then

-1.74

to

-1.59

 cuando

-1.31

.,,

-1.30

 ulter

-1.26

 ペア

-1.23

by

-1.23

 than

-1.21

Sì

-1.19

要用

-1.17

POSITIVE LOGITS

áveis

1.34

putar

1.30

The

1.30

 originalen

1.23

rógeno

1.23

 continuamente

1.23

 roupa

1.22

嗝

1.20

 entret

1.20

roskop

1.20

Activations Density 0.093%