INDEX

Explanations

opening quotation mark followed by sentence

The neuron strongly fires on opening quotation marks (and the start of quoted speech), i.e. it detects the beginnings of direct‐speech or quoted text.

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

琹

-1.53

มากๆ

-1.48

tements

-1.45

gants

-1.41

 esser

-1.40

 植物

-1.39

 have

-1.38

 parlato

-1.38

 hadde

-1.38

 suon

-1.38

POSITIVE LOGITS

).

1.64

 erneut

1.37

 contará

1.33

了个

1.28

🪦

1.27

 ebenso

1.26

が多く

1.25

 continuará

1.25

才知道

1.24

 comenzará

1.24

Activations Density 0.002%