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

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

Neuron 1613 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1536. Min Act: -0.0000

Data Index: 649159 (C4 (Web Text))

Max Activating Token Index: 320

Text #1

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1602. Min Act: -0.0000

Data Index: 989032 (C4 (Web Text))

Max Activating Token Index: 178

Text #2

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1507. Min Act: -0.0001

Data Index: 160932 (C4 (Web Text))

Max Activating Token Index: 353

Text #3

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1533. Min Act: -0.0001

Data Index: 1646900 (Python Code)

Max Activating Token Index: 10

Text #4

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1505. Min Act: -0.0001

Data Index: 924718 (C4 (Web Text))

Max Activating Token Index: 153

Text #5

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1338. Min Act: -0.0001

Data Index: 658339 (C4 (Web Text))

Max Activating Token Index: 548

Text #6

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1574. Min Act: -0.0001

Data Index: 1137968 (C4 (Web Text))

Max Activating Token Index: 536

Text #7

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1502. Min Act: -0.0001

Data Index: 1637356 (Python Code)

Max Activating Token Index: 30

Text #8

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1427. Min Act: -0.0000

Data Index: 1122532 (C4 (Web Text))

Max Activating Token Index: 396

Text #9

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1116. Min Act: -0.0000

Data Index: 32753 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1458. Min Act: -0.0000

Data Index: 1265689 (C4 (Web Text))

Max Activating Token Index: 157

Text #11

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1315. Min Act: -0.0000

Data Index: 463542 (C4 (Web Text))

Max Activating Token Index: 548

Text #12

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1351. Min Act: -0.0001

Data Index: 1417172 (Python Code)

Max Activating Token Index: 622

Text #13

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1326. Min Act: -0.0001

Data Index: 1469675 (Python Code)

Max Activating Token Index: 745

Text #14

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1376. Min Act: -0.0001

Data Index: 98691 (C4 (Web Text))

Max Activating Token Index: 86

Text #15

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1160. Min Act: -0.0001

Data Index: 628579 (C4 (Web Text))

Max Activating Token Index: 773

Text #16

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1273. Min Act: -0.0001

Data Index: 1494810 (Python Code)

Max Activating Token Index: 323

Text #17

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1223. Min Act: -0.0001

Data Index: 1467906 (Python Code)

Max Activating Token Index: 183

Text #18

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1211. Min Act: -0.0001

Data Index: 1063752 (C4 (Web Text))

Max Activating Token Index: 745

Text #19

Max Range: 0.1602. Min Range: -0.1602

Max Act: 0.1293. Min Act: -0.0001

Data Index: 1567833 (Python Code)

Max Activating Token Index: 17