INDEX

Explanations

coastline and neckline

The neuron activates on words ending in the suffix “-line.”

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

 жовтня

-2.81

ྥ

-2.45

 березня

-2.38

ा

-2.34

);

-2.30

🥂

-2.30

 вересня

-2.25

 липня

-2.23

出發

-2.22

POSITIVE LOGITS

 astute

2.83

2.69

蹰

2.63

 brazen

2.59

 fervent

2.58

2.53

 mitigating

2.42

 merciless

2.38

 catapult

2.33

 glittering

2.31

Activations Density 0.001%