INDEX

Explanations

numbers and punctuation

The neuron fires on bibliographic citation elements—journal or publisher names and their associated volume, year, and page-number figures.

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

 brainchild

0.38

 unmistakable

0.33

 nuestro

0.32

ۡ

0.32

 nosso

0.32

лад

0.31

ியின்

0.31

দের

0.30

 все

0.30

🔥🔥

0.30

POSITIVE LOGITS

,\

0.45

,),

0.42

pp

0.41

.),

0.41

.).

0.38

,)

0.38

 Ibid

0.38

).\

0.38

;\

0.37

,<

0.37

Activations Density 0.089%