INDEX

Explanations

latex mathematical formulas

The neuron selectively lights up on Shannon‐entropy style formula terms—especially the pᵢ and log pᵢ pieces of an entropy expression.

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

毆

-1.49

If

-1.41

 ataupun

-1.37

斯的

-1.36

 them

-1.35



-1.35

}$

-1.34

 genauer

-1.32

軀

-1.32

POSITIVE LOGITS

焢

1.40

 nyní

1.34

นี้

1.28

Gön

1.26

 artık

1.25

 теперь

1.25

 müsse

1.24

像

1.24

of

1.23

 permitan

1.23

Activations Density 0.012%