INDEX

Explanations

mathematical operators and symbols

This neuron activates on math operator tokens—especially differential‐operator names like “Delta,” “nabla,” and “gradient.”

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

-1.70

-1.55

-1.44

を取る

-1.41

跑到

-1.36

effec

-1.35

-1.34

-1.32

 ضبط

-1.32

POSITIVE LOGITS

’

1.50

ſelves

1.36

</strong>

1.35

</h2>

1.33

 customise

1.33

atzung

1.32

endif

1.32

!”

1.30

生の

1.30

 bigger

1.27

Activations Density 0.032%