INDEX

Explanations

common prepositions followed by various words

The neuron responds to qualifier and degree-expressing words (hedges, intensifiers, and other qualifying terms such as “not,” “usually,” “among,” “least,” 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

iações

-1.30

мѣ

-1.26

oldt

-1.21

ari

-1.20

pim

-1.20

怼

-1.16

籏

-1.16

ail

-1.13

嚆

-1.13

POSITIVE LOGITS

 people

1.55

encre

1.45

変わる

1.33

 arşivlendi

1.33

 höl

1.31

här

1.30

 geschenk

1.28

fabri

1.27

 which

1.26

 geheim

1.23

Activations Density 0.126%