INDEX

Explanations

gen followed by ch, _, k, too

The neuron flags occurrences of the character sequence “gen” (as in domain names like “gen.ch.edu.cn” or identifiers like “gen_server”).

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

↵↵↵

-1.55

mappedBy

-1.55

ῃ

-1.50

 Trabalho

-1.48

hanced

-1.45

皷

-1.41



-1.40

xvi

-1.40

 prohibido

-1.39



-1.37

POSITIVE LOGITS

1.84

1.60

1.57

 chied

1.48

1.46

1.45

with

1.43

of

1.41

any

1.40

my

1.39

Activations Density 0.015%