INDEX

Explanations

plugins

This neuron fires on code tokens referring to “plugin” (and “plugins”), i.e. occurrences of plugin-related identifiers and keys in code/config snippets.

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

 deuxième

-1.22

-1.17

🌸

-1.16

替代

-1.16

all

-1.15

 because

-1.14

 verbess

-1.13

oooooo

-1.13

adalajara

-1.12

ooooooo

-1.12

POSITIVE LOGITS

–

1.62

などは

1.55

 ceaseless

1.52

észetes

1.48

remia

1.43

 unceasing

1.39

1.32

eteria

1.31

uldron

1.30

 iaitu

1.29

Activations Density 0.268%