INDEX

Explanations

ordinal numbers followed by events

The neuron activates on ordinal number mentions (e.g. “first,” “second,” “third,” etc.), highlighting positions or rankings in lists.

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

 your

-1.34

 Mitarbeitern

-1.34

有很多

-1.34

 řízení

-1.34

 prevenir

-1.33

ziej

-1.31

wska

-1.23

 Menschheit

-1.23

 också

-1.21

 destruir

-1.20

POSITIVE LOGITS

of

1.71

other

1.58

for

1.48

нути

1.28

ījum

1.22

 other

1.20

 обучение

1.19

йлер

1.19

ια

1.16

ADDITIONAL

1.16

Activations Density 0.040%