Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3690 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1825. Min Act: -0.0001

Data Index: 215756 (C4 (Web Text))

Max Activating Token Index: 105

Text #1

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1607. Min Act: -0.0001

Data Index: 694510 (C4 (Web Text))

Max Activating Token Index: 836

Text #2

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1551. Min Act: -0.0001

Data Index: 356896 (C4 (Web Text))

Max Activating Token Index: 256

Text #3

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1458. Min Act: -0.0001

Data Index: 1022812 (C4 (Web Text))

Max Activating Token Index: 10

Text #4

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1395. Min Act: -0.0001

Data Index: 207405 (C4 (Web Text))

Max Activating Token Index: 787

Text #5

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1373. Min Act: -0.0001

Data Index: 88577 (C4 (Web Text))

Max Activating Token Index: 40

Text #6

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1425. Min Act: -0.0000

Data Index: 1604501 (Python Code)

Max Activating Token Index: 838

Text #7

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1410. Min Act: -0.0001

Data Index: 589741 (C4 (Web Text))

Max Activating Token Index: 583

Text #8

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1402. Min Act: -0.0001

Data Index: 266595 (C4 (Web Text))

Max Activating Token Index: 433

Text #9

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1348. Min Act: -0.0001

Data Index: 367171 (C4 (Web Text))

Max Activating Token Index: 834

Text #10

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1294. Min Act: -0.0001

Data Index: 965108 (C4 (Web Text))

Max Activating Token Index: 389

Text #11

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1333. Min Act: -0.0001

Data Index: 680796 (C4 (Web Text))

Max Activating Token Index: 659

Text #12

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1240. Min Act: -0.0001

Data Index: 1422568 (Python Code)

Max Activating Token Index: 215

Text #13

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1390. Min Act: -0.0001

Data Index: 1244827 (C4 (Web Text))

Max Activating Token Index: 196

Text #14

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1331. Min Act: -0.0001

Data Index: 176317 (C4 (Web Text))

Max Activating Token Index: 51

Text #15

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1336. Min Act: -0.0001

Data Index: 920408 (C4 (Web Text))

Max Activating Token Index: 129

Text #16

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1327. Min Act: -0.0001

Data Index: 149268 (C4 (Web Text))

Max Activating Token Index: 289

Text #17

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1272. Min Act: -0.0001

Data Index: 484399 (C4 (Web Text))

Max Activating Token Index: 165

Text #18

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1259. Min Act: -0.0001

Data Index: 565753 (C4 (Web Text))

Max Activating Token Index: 238

Text #19

Max Range: 0.1825. Min Range: -0.1825

Max Act: 0.1304. Min Act: -0.0001

Data Index: 565617 (C4 (Web Text))

Max Activating Token Index: 510