INDEX

Explanations

listing reasons and purposes

The neuron activates on discourse markers that introduce enumerations or lists (e.g. “purposes,” “reasons,” “factors,” “two-fold,” etc.).

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

褙

-1.30

黽

-1.28

或者

-1.28

 version

-1.23

nością

-1.21

hythm

-1.20

ِّ

-1.16

 Prefeitura

-1.15

爰

-1.15

acidade

-1.14

POSITIVE LOGITS

</em>

1.43

1.39

 seyn

1.33

but

1.32

 verschiedenen

1.24

 things

1.22

 both

1.21

viu

1.17

 olika

1.15

Về

1.15

Activations Density 0.163%