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

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

Neuron 1720 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.3632. Min Act: -0.0001

Data Index: 16039 (C4 (Web Text))

Max Activating Token Index: 996

Text #1

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.2407. Min Act: -0.0001

Data Index: 729295 (C4 (Web Text))

Max Activating Token Index: 481

Text #2

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.3570. Min Act: -0.0000

Data Index: 410769 (C4 (Web Text))

Max Activating Token Index: 425

Text #3

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.0907. Min Act: -0.0001

Data Index: 953049 (C4 (Web Text))

Max Activating Token Index: 301

Text #4

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.1066. Min Act: -0.0000

Data Index: 765298 (C4 (Web Text))

Max Activating Token Index: 329

Text #5

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.0762. Min Act: -0.0001

Data Index: 1241084 (C4 (Web Text))

Max Activating Token Index: 332

Text #6

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.1848. Min Act: -0.0001

Data Index: 272741 (C4 (Web Text))

Max Activating Token Index: 1009

Text #7

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.9039. Min Act: -0.0000

Data Index: 516604 (C4 (Web Text))

Max Activating Token Index: 834

Text #8

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.0114. Min Act: -0.0000

Data Index: 881468 (C4 (Web Text))

Max Activating Token Index: 64

Text #9

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.0144. Min Act: -0.0001

Data Index: 1060970 (C4 (Web Text))

Max Activating Token Index: 716

Text #10

Max Range: 3.3632. Min Range: -3.3632

Max Act: 3.2546. Min Act: -0.0001

Data Index: 558595 (C4 (Web Text))

Max Activating Token Index: 579

Text #11

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.5508. Min Act: -0.0000

Data Index: 977707 (C4 (Web Text))

Max Activating Token Index: 965

Text #12

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.7952. Min Act: -0.0000

Data Index: 1067571 (C4 (Web Text))

Max Activating Token Index: 378

Text #13

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.9239. Min Act: -0.0001

Data Index: 349320 (C4 (Web Text))

Max Activating Token Index: 639

Text #14

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.7565. Min Act: -0.0001

Data Index: 271985 (C4 (Web Text))

Max Activating Token Index: 199

Text #15

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.8832. Min Act: -0.0000

Data Index: 720500 (C4 (Web Text))

Max Activating Token Index: 714

Text #16

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.9172. Min Act: -0.0000

Data Index: 1365721 (Python Code)

Max Activating Token Index: 790

Text #17

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.7332. Min Act: -0.0000

Data Index: 362557 (C4 (Web Text))

Max Activating Token Index: 662

Text #18

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.9822. Min Act: -0.0001

Data Index: 530646 (C4 (Web Text))

Max Activating Token Index: 94

Text #19

Max Range: 3.3632. Min Range: -3.3632

Max Act: 2.8555. Min Act: -0.0000

Data Index: 849509 (C4 (Web Text))

Max Activating Token Index: 519