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

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

Neuron 351 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.4544. Min Act: -0.0001

Data Index: 1474574 (Python Code)

Max Activating Token Index: 5

Text #1

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.4173. Min Act: -0.0001

Data Index: 1310615 (C4 (Web Text))

Max Activating Token Index: 727

Text #2

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3721. Min Act: -0.0001

Data Index: 1124566 (C4 (Web Text))

Max Activating Token Index: 936

Text #3

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3732. Min Act: -0.0002

Data Index: 178929 (C4 (Web Text))

Max Activating Token Index: 475

Text #4

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3824. Min Act: -0.0002

Data Index: 883726 (C4 (Web Text))

Max Activating Token Index: 14

Text #5

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3741. Min Act: -0.0002

Data Index: 673666 (C4 (Web Text))

Max Activating Token Index: 363

Text #6

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3268. Min Act: -0.0001

Data Index: 854630 (C4 (Web Text))

Max Activating Token Index: 932

Text #7

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3698. Min Act: -0.0001

Data Index: 1314622 (C4 (Web Text))

Max Activating Token Index: 10

Text #8

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3596. Min Act: -0.0002

Data Index: 460308 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3705. Min Act: -0.0001

Data Index: 665352 (C4 (Web Text))

Max Activating Token Index: 939

Text #10

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3585. Min Act: -0.0002

Data Index: 368896 (C4 (Web Text))

Max Activating Token Index: 796

Text #11

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3570. Min Act: -0.0001

Data Index: 1014537 (C4 (Web Text))

Max Activating Token Index: 397

Text #12

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3028. Min Act: -0.0001

Data Index: 1458372 (Python Code)

Max Activating Token Index: 755

Text #13

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3281. Min Act: -0.0001

Data Index: 726165 (C4 (Web Text))

Max Activating Token Index: 757

Text #14

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3085. Min Act: -0.0001

Data Index: 21385 (C4 (Web Text))

Max Activating Token Index: 939

Text #15

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3247. Min Act: -0.0001

Data Index: 120917 (C4 (Web Text))

Max Activating Token Index: 44

Text #16

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3277. Min Act: -0.0001

Data Index: 1141817 (C4 (Web Text))

Max Activating Token Index: 72

Text #17

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3379. Min Act: -0.0002

Data Index: 825977 (C4 (Web Text))

Max Activating Token Index: 18

Text #18

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3478. Min Act: -0.0001

Data Index: 502571 (C4 (Web Text))

Max Activating Token Index: 427

Text #19

Max Range: 0.4544. Min Range: -0.4544

Max Act: 0.3423. Min Act: -0.0001

Data Index: 1516945 (Python Code)

Max Activating Token Index: 966