INDEX

Explanations

chemical names and formulas

This neuron detects chemical formulas and structural notations, particularly those with numbers, hyphens, and parentheses in chemical nomenclature.

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

)):

-1.80

if

-1.64

 even

-1.64

 bahkan

-1.63

 حتی

-1.60

"]();

-1.58

})$,

-1.55

 nawet

-1.54

")));

-1.53

even

-1.50

POSITIVE LOGITS

brakk

1.37

 streamlined

1.30

 débit

1.28

 slav

1.26

 joueurs

1.23

俢

1.23

jauh

1.22

 hôtes

1.19

Sneaky

1.19

Activations Density 0.078%