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

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

Neuron 656 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.1791. Min Range: -9.1791

Max Act: 9.1791. Min Act: -0.0001

Data Index: 1242167 (C4 (Web Text))

Max Activating Token Index: 353

Text #1

Max Range: 9.1791. Min Range: -9.1791

Max Act: 9.0113. Min Act: -0.0001

Data Index: 104700 (C4 (Web Text))

Max Activating Token Index: 176

Text #2

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.9769. Min Act: -0.0001

Data Index: 402621 (C4 (Web Text))

Max Activating Token Index: 458

Text #3

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.6414. Min Act: -0.0001

Data Index: 1195392 (C4 (Web Text))

Max Activating Token Index: 616

Text #4

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.7684. Min Act: -0.0001

Data Index: 1053555 (C4 (Web Text))

Max Activating Token Index: 684

Text #5

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.7352. Min Act: -0.0001

Data Index: 755015 (C4 (Web Text))

Max Activating Token Index: 476

Text #6

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.9611. Min Act: -0.0001

Data Index: 814845 (C4 (Web Text))

Max Activating Token Index: 866

Text #7

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.9469. Min Act: -0.0001

Data Index: 884089 (C4 (Web Text))

Max Activating Token Index: 737

Text #8

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.7915. Min Act: -0.0001

Data Index: 1317168 (C4 (Web Text))

Max Activating Token Index: 566

Text #9

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.8080. Min Act: -0.0001

Data Index: 1285630 (C4 (Web Text))

Max Activating Token Index: 349

Text #10

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.9041. Min Act: -0.0001

Data Index: 966066 (C4 (Web Text))

Max Activating Token Index: 334

Text #11

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.2512. Min Act: -0.0001

Data Index: 155625 (C4 (Web Text))

Max Activating Token Index: 814

Text #12

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.6879. Min Act: -0.0001

Data Index: 258300 (C4 (Web Text))

Max Activating Token Index: 237

Text #13

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.6905. Min Act: -0.0001

Data Index: 1766 (C4 (Web Text))

Max Activating Token Index: 478

Text #14

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.2340. Min Act: -0.0001

Data Index: 774712 (C4 (Web Text))

Max Activating Token Index: 603

Text #15

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.5067. Min Act: -0.0001

Data Index: 367041 (C4 (Web Text))

Max Activating Token Index: 410

Text #16

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.4990. Min Act: -0.0001

Data Index: 1071285 (C4 (Web Text))

Max Activating Token Index: 464

Text #17

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.4812. Min Act: -0.0001

Data Index: 788914 (C4 (Web Text))

Max Activating Token Index: 150

Text #18

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.9050. Min Act: -0.0001

Data Index: 1221554 (C4 (Web Text))

Max Activating Token Index: 138

Text #19

Max Range: 9.1791. Min Range: -9.1791

Max Act: 8.6167. Min Act: -0.0001

Data Index: 1011279 (C4 (Web Text))

Max Activating Token Index: 136