Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2391 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1787. Min Act: -0.0001

Data Index: 384520 (C4 (Web Text))

Max Activating Token Index: 257

Text #1

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1800. Min Act: -0.0001

Data Index: 33675 (C4 (Web Text))

Max Activating Token Index: 188

Text #2

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1701. Min Act: -0.0001

Data Index: 527762 (C4 (Web Text))

Max Activating Token Index: 322

Text #3

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1383. Min Act: -0.0001

Data Index: 1622669 (Python Code)

Max Activating Token Index: 288

Text #4

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1541. Min Act: -0.0001

Data Index: 588315 (C4 (Web Text))

Max Activating Token Index: 48

Text #5

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1855. Min Act: -0.0001

Data Index: 1014637 (C4 (Web Text))

Max Activating Token Index: 349

Text #6

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1383. Min Act: -0.0001

Data Index: 1220341 (C4 (Web Text))

Max Activating Token Index: 278

Text #7

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1588. Min Act: -0.0001

Data Index: 831608 (C4 (Web Text))

Max Activating Token Index: 459

Text #8

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1516. Min Act: -0.0001

Data Index: 491454 (C4 (Web Text))

Max Activating Token Index: 387

Text #9

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1269. Min Act: -0.0001

Data Index: 1545851 (Python Code)

Max Activating Token Index: 609

Text #10

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1358. Min Act: -0.0001

Data Index: 685419 (C4 (Web Text))

Max Activating Token Index: 636

Text #11

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1572. Min Act: -0.0001

Data Index: 1544149 (Python Code)

Max Activating Token Index: 404

Text #12

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1429. Min Act: -0.0001

Data Index: 1559001 (Python Code)

Max Activating Token Index: 601

Text #13

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1135. Min Act: -0.0001

Data Index: 225008 (C4 (Web Text))

Max Activating Token Index: 771

Text #14

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1331. Min Act: -0.0001

Data Index: 1545490 (Python Code)

Max Activating Token Index: 274

Text #15

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1616. Min Act: -0.0001

Data Index: 975866 (C4 (Web Text))

Max Activating Token Index: 157

Text #16

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1512. Min Act: -0.0001

Data Index: 1084379 (C4 (Web Text))

Max Activating Token Index: 785

Text #17

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1037. Min Act: -0.0001

Data Index: 838850 (C4 (Web Text))

Max Activating Token Index: 736

Text #18

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1530. Min Act: -0.0001

Data Index: 1128820 (C4 (Web Text))

Max Activating Token Index: 65

Text #19

Max Range: 0.1855. Min Range: -0.1855

Max Act: 0.1398. Min Act: -0.0001

Data Index: 787550 (C4 (Web Text))

Max Activating Token Index: 663