INDEX

Explanations

uninstall, unsubscribe, unpublish

The neuron activates strongly on “un-” prefixed words denoting removal or reversal (e.g. uninstall, uninstaller, unmap).

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

 sebelum

-2.25

 standout

-2.13

 dazzling

-2.11

 Maybe

-2.06

 staunch

-2.05

وانید

-1.85

 Sadly

-1.80

 інші

-1.79

 Usually

-1.77

 усі

-1.76

POSITIVE LOGITS

🛇

2.42

᯾

2.08

妡

2.02

thodoxy

2.00

𖧵

1.96

萏

1.95

麻辣

1.93

嗷

1.90

for

1.90

媜

1.86

Activations Density 0.008%