INDEX

Explanations

contracted prevention handled with past

The neuron fires strongly on tokens that are parts of people’s names (proper‐name entities).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 zweimal

0.37

 zunächst

0.36

 stopwatch

0.36

 delightfully

0.35

 двумя

0.35

 Stopwatch

0.34

 featuring

0.34

 nettement

0.34

 majoring

0.34

 dvě

0.33

POSITIVE LOGITS

 organisms

0.43

 people

0.42

 pessoas

0.42

 businesses

0.42

人们

0.42

人們

0.42

 insanlar

0.40

businesses

0.40

 ludzi

0.40

 entreprises

0.39

Activations Density 0.006%