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

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

Neuron 1044 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.3370. Min Act: -0.0000

Data Index: 655854 (C4 (Web Text))

Max Activating Token Index: 810

Text #1

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2718. Min Act: -0.0000

Data Index: 1329550 (C4 (Web Text))

Max Activating Token Index: 115

Text #2

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2462. Min Act: -0.0000

Data Index: 1388670 (Python Code)

Max Activating Token Index: 51

Text #3

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2445. Min Act: -0.0000

Data Index: 307648 (C4 (Web Text))

Max Activating Token Index: 869

Text #4

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2406. Min Act: -0.0000

Data Index: 1207421 (C4 (Web Text))

Max Activating Token Index: 968

Text #5

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2512. Min Act: -0.0000

Data Index: 23311 (C4 (Web Text))

Max Activating Token Index: 671

Text #6

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2332. Min Act: -0.0000

Data Index: 1390785 (Python Code)

Max Activating Token Index: 54

Text #7

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2307. Min Act: -0.0000

Data Index: 307627 (C4 (Web Text))

Max Activating Token Index: 849

Text #8

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2455. Min Act: -0.0000

Data Index: 1022426 (C4 (Web Text))

Max Activating Token Index: 488

Text #9

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2385. Min Act: -0.0000

Data Index: 8636 (C4 (Web Text))

Max Activating Token Index: 861

Text #10

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2453. Min Act: -0.0000

Data Index: 995044 (C4 (Web Text))

Max Activating Token Index: 407

Text #11

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2451. Min Act: -0.0000

Data Index: 918513 (C4 (Web Text))

Max Activating Token Index: 105

Text #12

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2526. Min Act: -0.0000

Data Index: 523382 (C4 (Web Text))

Max Activating Token Index: 613

Text #13

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2456. Min Act: -0.0000

Data Index: 1592096 (Python Code)

Max Activating Token Index: 41

Text #14

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2565. Min Act: -0.0000

Data Index: 381858 (C4 (Web Text))

Max Activating Token Index: 770

Text #15

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2290. Min Act: -0.0000

Data Index: 929053 (C4 (Web Text))

Max Activating Token Index: 685

Text #16

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2275. Min Act: -0.0000

Data Index: 547615 (C4 (Web Text))

Max Activating Token Index: 959

Text #17

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2365. Min Act: -0.0000

Data Index: 1495388 (Python Code)

Max Activating Token Index: 39

Text #18

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2322. Min Act: -0.0000

Data Index: 1074367 (C4 (Web Text))

Max Activating Token Index: 474

Text #19

Max Range: 0.3370. Min Range: -0.3370

Max Act: 0.2408. Min Act: -0.0000

Data Index: 45667 (C4 (Web Text))

Max Activating Token Index: 596