Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1556 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.9452. Min Act: -0.0006

Data Index: 513706 (C4 (Web Text))

Max Activating Token Index: 852

Text #1

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.8384. Min Act: -0.0006

Data Index: 1200273 (C4 (Web Text))

Max Activating Token Index: 378

Text #2

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.9677. Min Act: -0.0006

Data Index: 1187162 (C4 (Web Text))

Max Activating Token Index: 684

Text #3

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7844. Min Act: -0.0006

Data Index: 1031400 (C4 (Web Text))

Max Activating Token Index: 85

Text #4

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7881. Min Act: -0.0006

Data Index: 426645 (C4 (Web Text))

Max Activating Token Index: 184

Text #5

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7789. Min Act: -0.0006

Data Index: 1310254 (C4 (Web Text))

Max Activating Token Index: 9

Text #6

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.8379. Min Act: -0.0006

Data Index: 1248387 (C4 (Web Text))

Max Activating Token Index: 644

Text #7

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7333. Min Act: -0.0006

Data Index: 1047207 (C4 (Web Text))

Max Activating Token Index: 87

Text #8

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7395. Min Act: -0.0006

Data Index: 823165 (C4 (Web Text))

Max Activating Token Index: 214

Text #9

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7881. Min Act: -0.0006

Data Index: 237388 (C4 (Web Text))

Max Activating Token Index: 810

Text #10

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7364. Min Act: -0.0006

Data Index: 580614 (C4 (Web Text))

Max Activating Token Index: 475

Text #11

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.6835. Min Act: -0.0006

Data Index: 1005782 (C4 (Web Text))

Max Activating Token Index: 329

Text #12

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7664. Min Act: -0.0006

Data Index: 547029 (C4 (Web Text))

Max Activating Token Index: 203

Text #13

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7618. Min Act: -0.0006

Data Index: 1181090 (C4 (Web Text))

Max Activating Token Index: 41

Text #14

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.6412. Min Act: -0.0006

Data Index: 1178131 (C4 (Web Text))

Max Activating Token Index: 941

Text #15

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.8067. Min Act: -0.0006

Data Index: 835897 (C4 (Web Text))

Max Activating Token Index: 644

Text #16

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.6159. Min Act: -0.0006

Data Index: 1030431 (C4 (Web Text))

Max Activating Token Index: 845

Text #17

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.6810. Min Act: -0.0006

Data Index: 44302 (C4 (Web Text))

Max Activating Token Index: 633

Text #18

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.7023. Min Act: -0.0006

Data Index: 1357689 (C4 (Web Text))

Max Activating Token Index: 764

Text #19

Max Range: 0.9677. Min Range: -0.9677

Max Act: 0.6549. Min Act: -0.0006

Data Index: 105962 (C4 (Web Text))

Max Activating Token Index: 807