Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3000 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.2292. Min Act: -0.0001

Data Index: 297472 (C4 (Web Text))

Max Activating Token Index: 525

Text #1

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.2088. Min Act: -0.0001

Data Index: 460947 (C4 (Web Text))

Max Activating Token Index: 157

Text #2

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1910. Min Act: -0.0001

Data Index: 813461 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1960. Min Act: -0.0001

Data Index: 424948 (C4 (Web Text))

Max Activating Token Index: 845

Text #4

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1684. Min Act: -0.0001

Data Index: 38429 (C4 (Web Text))

Max Activating Token Index: 810

Text #5

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1786. Min Act: -0.0001

Data Index: 532346 (C4 (Web Text))

Max Activating Token Index: 289

Text #6

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1793. Min Act: -0.0001

Data Index: 877917 (C4 (Web Text))

Max Activating Token Index: 303

Text #7

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1747. Min Act: -0.0001

Data Index: 1322123 (C4 (Web Text))

Max Activating Token Index: 683

Text #8

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1625. Min Act: -0.0001

Data Index: 63296 (C4 (Web Text))

Max Activating Token Index: 815

Text #9

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1793. Min Act: -0.0001

Data Index: 1133375 (C4 (Web Text))

Max Activating Token Index: 56

Text #10

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1652. Min Act: -0.0001

Data Index: 691423 (C4 (Web Text))

Max Activating Token Index: 590

Text #11

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1601. Min Act: -0.0001

Data Index: 689165 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1531. Min Act: -0.0001

Data Index: 998370 (C4 (Web Text))

Max Activating Token Index: 611

Text #13

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1626. Min Act: -0.0001

Data Index: 281396 (C4 (Web Text))

Max Activating Token Index: 374

Text #14

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1661. Min Act: -0.0001

Data Index: 715432 (C4 (Web Text))

Max Activating Token Index: 143

Text #15

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1641. Min Act: -0.0001

Data Index: 417028 (C4 (Web Text))

Max Activating Token Index: 831

Text #16

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1630. Min Act: -0.0001

Data Index: 1546153 (Python Code)

Max Activating Token Index: 392

Text #17

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1622. Min Act: -0.0001

Data Index: 475755 (C4 (Web Text))

Max Activating Token Index: 429

Text #18

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1611. Min Act: -0.0001

Data Index: 317197 (C4 (Web Text))

Max Activating Token Index: 243

Text #19

Max Range: 0.2292. Min Range: -0.2292

Max Act: 0.1525. Min Act: -0.0001

Data Index: 128604 (C4 (Web Text))

Max Activating Token Index: 613