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

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

Neuron 2953 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.9443. Min Range: -4.9443

Max Act: 4.9443. Min Act: -0.0000

Data Index: 1293683 (C4 (Web Text))

Max Activating Token Index: 876

Text #1

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.9173. Min Act: -0.0000

Data Index: 666517 (C4 (Web Text))

Max Activating Token Index: 594

Text #2

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.8724. Min Act: -0.0000

Data Index: 233140 (C4 (Web Text))

Max Activating Token Index: 576

Text #3

Max Range: 4.9443. Min Range: -4.9443

Max Act: 4.0521. Min Act: -0.0000

Data Index: 1125704 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.9490. Min Act: 0.0000

Data Index: 733525 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.9361. Min Act: -0.0000

Data Index: 1483002 (Python Code)

Max Activating Token Index: 24

Text #6

Max Range: 4.9443. Min Range: -4.9443

Max Act: 2.9586. Min Act: -0.0000

Data Index: 1075902 (C4 (Web Text))

Max Activating Token Index: 880

Text #7

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.3217. Min Act: -0.0000

Data Index: 668478 (C4 (Web Text))

Max Activating Token Index: 862

Text #8

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.1695. Min Act: 0.0000

Data Index: 113326 (C4 (Web Text))

Max Activating Token Index: 457

Text #9

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.0185. Min Act: -0.0000

Data Index: 1398051 (Python Code)

Max Activating Token Index: 622

Text #10

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.2501. Min Act: -0.0000

Data Index: 1011171 (C4 (Web Text))

Max Activating Token Index: 661

Text #11

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.4125. Min Act: -0.0000

Data Index: 732722 (C4 (Web Text))

Max Activating Token Index: 652

Text #12

Max Range: 4.9443. Min Range: -4.9443

Max Act: 2.9706. Min Act: -0.0000

Data Index: 1618351 (Python Code)

Max Activating Token Index: 525

Text #13

Max Range: 4.9443. Min Range: -4.9443

Max Act: 2.9655. Min Act: -0.0000

Data Index: 17412 (C4 (Web Text))

Max Activating Token Index: 423

Text #14

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.1005. Min Act: -0.0000

Data Index: 1235461 (C4 (Web Text))

Max Activating Token Index: 537

Text #15

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.3375. Min Act: -0.0000

Data Index: 1214992 (C4 (Web Text))

Max Activating Token Index: 871

Text #16

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.0065. Min Act: -0.0000

Data Index: 1034610 (C4 (Web Text))

Max Activating Token Index: 646

Text #17

Max Range: 4.9443. Min Range: -4.9443

Max Act: 3.0297. Min Act: -0.0000

Data Index: 1075100 (C4 (Web Text))

Max Activating Token Index: 499

Text #18

Max Range: 4.9443. Min Range: -4.9443

Max Act: 2.7132. Min Act: -0.0000

Data Index: 95314 (C4 (Web Text))

Max Activating Token Index: 720

Text #19

Max Range: 4.9443. Min Range: -4.9443

Max Act: 2.7306. Min Act: -0.0000

Data Index: 618422 (C4 (Web Text))

Max Activating Token Index: 473