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

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

Neuron 3629 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: 0.4420. Min Range: -0.4420

Max Act: 0.4420. Min Act: -0.0002

Data Index: 1285539 (C4 (Web Text))

Max Activating Token Index: 797

Text #1

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.4299. Min Act: -0.0001

Data Index: 1609603 (Python Code)

Max Activating Token Index: 949

Text #2

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.4351. Min Act: -0.0001

Data Index: 1470106 (Python Code)

Max Activating Token Index: 8

Text #3

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3848. Min Act: -0.0002

Data Index: 964632 (C4 (Web Text))

Max Activating Token Index: 1001

Text #4

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3686. Min Act: -0.0002

Data Index: 1271168 (C4 (Web Text))

Max Activating Token Index: 963

Text #5

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3626. Min Act: -0.0001

Data Index: 1411912 (Python Code)

Max Activating Token Index: 10

Text #6

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3414. Min Act: -0.0001

Data Index: 1617874 (Python Code)

Max Activating Token Index: 506

Text #7

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3810. Min Act: -0.0001

Data Index: 1538065 (Python Code)

Max Activating Token Index: 9

Text #8

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3297. Min Act: -0.0002

Data Index: 524014 (C4 (Web Text))

Max Activating Token Index: 981

Text #9

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3732. Min Act: -0.0001

Data Index: 1476537 (Python Code)

Max Activating Token Index: 23

Text #10

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3754. Min Act: -0.0002

Data Index: 1305737 (C4 (Web Text))

Max Activating Token Index: 986

Text #11

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3586. Min Act: -0.0002

Data Index: 135208 (C4 (Web Text))

Max Activating Token Index: 890

Text #12

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3722. Min Act: -0.0001

Data Index: 842174 (C4 (Web Text))

Max Activating Token Index: 833

Text #13

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3183. Min Act: -0.0002

Data Index: 936619 (C4 (Web Text))

Max Activating Token Index: 179

Text #14

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.2946. Min Act: -0.0002

Data Index: 755958 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3564. Min Act: -0.0001

Data Index: 1614080 (Python Code)

Max Activating Token Index: 880

Text #16

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.2824. Min Act: -0.0001

Data Index: 850386 (C4 (Web Text))

Max Activating Token Index: 878

Text #17

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3678. Min Act: -0.0001

Data Index: 1561592 (Python Code)

Max Activating Token Index: 77

Text #18

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3363. Min Act: -0.0001

Data Index: 868350 (C4 (Web Text))

Max Activating Token Index: 558

Text #19

Max Range: 0.4420. Min Range: -0.4420

Max Act: 0.3334. Min Act: -0.0001

Data Index: 79 (C4 (Web Text))

Max Activating Token Index: 222