INDEX

Explanations

AI related terms

The neuron fires on assertions framing artificial intelligence as an existential or world‐ending threat to humanity.

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

all

-1.37

十八

-1.29

ଂ

-1.27

for

-1.26

 sezon

-1.25

Ḵ

-1.24

 longstanding

-1.24

 трябва

-1.21

 generally

-1.20

 süd

-1.16

POSITIVE LOGITS

 pomocí

1.53

 ativar

1.49

 gerar

1.40

汌

1.38

 your

1.37

 executar

1.37

Джере

1.35

 její

1.34

慢慢的

1.32

 nové

1.31

Activations Density 0.147%