INDEX

Explanations

firstgameraceinterviewpost

This neuron fires on mentions of an inaugural or “first” occurrence—e.g. first game, first interview, first post—highlighting passages about someone’s or something’s debut.

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

</table>

-1.49

-1.40

 szok

-1.33

invokeLater

-1.33

“

-1.31

蚰

-1.29

 debemos

-1.29

ursing

-1.27

 alguna

-1.26

:])

-1.25

POSITIVE LOGITS

 from

1.92

 that

1.78

for

1.58

 after

1.44

his

1.41

of

1.36

 nowego

1.35

してしまった

1.29

 until

1.24

on

1.23

Activations Density 0.084%