Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 6020 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.4102. Min Range: -4.4102

Max Act: 4.4102. Min Act: -0.0001

Data Index: 1206789 (C4 (Web Text))

Max Activating Token Index: 889

Text #1

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.8701. Min Act: -0.0001

Data Index: 961419 (C4 (Web Text))

Max Activating Token Index: 904

Text #2

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.7740. Min Act: -0.0000

Data Index: 63687 (C4 (Web Text))

Max Activating Token Index: 812

Text #3

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.5686. Min Act: -0.0000

Data Index: 1487171 (Python Code)

Max Activating Token Index: 291

Text #4

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.6076. Min Act: -0.0001

Data Index: 988007 (C4 (Web Text))

Max Activating Token Index: 891

Text #5

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.4434. Min Act: -0.0001

Data Index: 1344658 (C4 (Web Text))

Max Activating Token Index: 757

Text #6

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.6737. Min Act: -0.0000

Data Index: 1220749 (C4 (Web Text))

Max Activating Token Index: 291

Text #7

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.5929. Min Act: -0.0001

Data Index: 1395132 (Python Code)

Max Activating Token Index: 235

Text #8

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.7551. Min Act: -0.0000

Data Index: 1234967 (C4 (Web Text))

Max Activating Token Index: 360

Text #9

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.7010. Min Act: -0.0000

Data Index: 759184 (C4 (Web Text))

Max Activating Token Index: 84

Text #10

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.8363. Min Act: -0.0001

Data Index: 1099231 (C4 (Web Text))

Max Activating Token Index: 716

Text #11

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.4425. Min Act: -0.0001

Data Index: 1063784 (C4 (Web Text))

Max Activating Token Index: 624

Text #12

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.0848. Min Act: -0.0001

Data Index: 750442 (C4 (Web Text))

Max Activating Token Index: 649

Text #13

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.5468. Min Act: -0.0000

Data Index: 683086 (C4 (Web Text))

Max Activating Token Index: 967

Text #14

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.2097. Min Act: -0.0001

Data Index: 1169441 (C4 (Web Text))

Max Activating Token Index: 726

Text #15

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.1974. Min Act: -0.0000

Data Index: 221732 (C4 (Web Text))

Max Activating Token Index: 492

Text #16

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.2927. Min Act: -0.0000

Data Index: 230305 (C4 (Web Text))

Max Activating Token Index: 187

Text #17

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.1418. Min Act: -0.0001

Data Index: 1582470 (Python Code)

Max Activating Token Index: 978

Text #18

Max Range: 4.4102. Min Range: -4.4102

Max Act: 3.3421. Min Act: -0.0000

Data Index: 432740 (C4 (Web Text))

Max Activating Token Index: 411

Text #19

Max Range: 4.4102. Min Range: -4.4102

Max Act: 2.7869. Min Act: -0.0000

Data Index: 703686 (C4 (Web Text))

Max Activating Token Index: 942