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

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

Neuron 338 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.9102. Min Act: -0.0003

Data Index: 199732 (C4 (Web Text))

Max Activating Token Index: 809

Text #1

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.6968. Min Act: -0.0003

Data Index: 207787 (C4 (Web Text))

Max Activating Token Index: 810

Text #2

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.6053. Min Act: -0.0003

Data Index: 1383993 (Python Code)

Max Activating Token Index: 529

Text #3

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.5374. Min Act: -0.0003

Data Index: 996505 (C4 (Web Text))

Max Activating Token Index: 707

Text #4

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.4445. Min Act: -0.0003

Data Index: 45948 (C4 (Web Text))

Max Activating Token Index: 708

Text #5

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.4668. Min Act: -0.0003

Data Index: 770137 (C4 (Web Text))

Max Activating Token Index: 611

Text #6

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3916. Min Act: -0.0003

Data Index: 496332 (C4 (Web Text))

Max Activating Token Index: 85

Text #7

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3874. Min Act: -0.0003

Data Index: 703180 (C4 (Web Text))

Max Activating Token Index: 862

Text #8

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3855. Min Act: -0.0003

Data Index: 163295 (C4 (Web Text))

Max Activating Token Index: 155

Text #9

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3051. Min Act: -0.0003

Data Index: 681161 (C4 (Web Text))

Max Activating Token Index: 517

Text #10

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3814. Min Act: -0.0003

Data Index: 1112223 (C4 (Web Text))

Max Activating Token Index: 70

Text #11

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3272. Min Act: -0.0003

Data Index: 1336284 (C4 (Web Text))

Max Activating Token Index: 864

Text #12

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3241. Min Act: -0.0003

Data Index: 1004724 (C4 (Web Text))

Max Activating Token Index: 652

Text #13

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3231. Min Act: -0.0003

Data Index: 1188848 (C4 (Web Text))

Max Activating Token Index: 817

Text #14

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3190. Min Act: -0.0003

Data Index: 525482 (C4 (Web Text))

Max Activating Token Index: 528

Text #15

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.3839. Min Act: -0.0003

Data Index: 893504 (C4 (Web Text))

Max Activating Token Index: 686

Text #16

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.2624. Min Act: -0.0003

Data Index: 1645851 (Python Code)

Max Activating Token Index: 922

Text #17

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.2431. Min Act: -0.0003

Data Index: 514430 (C4 (Web Text))

Max Activating Token Index: 414

Text #18

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.2405. Min Act: -0.0003

Data Index: 164358 (C4 (Web Text))

Max Activating Token Index: 835

Text #19

Max Range: 11.9102. Min Range: -11.9102

Max Act: 11.2384. Min Act: -0.0003

Data Index: 788775 (C4 (Web Text))

Max Activating Token Index: 1014