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

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

Neuron 3624 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6373. Min Act: -0.0002

Data Index: 461024 (C4 (Web Text))

Max Activating Token Index: 27

Text #1

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6949. Min Act: -0.0002

Data Index: 1459316 (Python Code)

Max Activating Token Index: 1003

Text #2

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6890. Min Act: -0.0002

Data Index: 636718 (C4 (Web Text))

Max Activating Token Index: 722

Text #3

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5827. Min Act: -0.0002

Data Index: 1032403 (C4 (Web Text))

Max Activating Token Index: 98

Text #4

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6765. Min Act: -0.0002

Data Index: 780018 (C4 (Web Text))

Max Activating Token Index: 657

Text #5

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5743. Min Act: -0.0002

Data Index: 603355 (C4 (Web Text))

Max Activating Token Index: 935

Text #6

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6102. Min Act: -0.0002

Data Index: 207871 (C4 (Web Text))

Max Activating Token Index: 44

Text #7

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5488. Min Act: -0.0002

Data Index: 1130809 (C4 (Web Text))

Max Activating Token Index: 862

Text #8

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5847. Min Act: -0.0002

Data Index: 59638 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5757. Min Act: -0.0002

Data Index: 716410 (C4 (Web Text))

Max Activating Token Index: 412

Text #10

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5613. Min Act: -0.0002

Data Index: 745414 (C4 (Web Text))

Max Activating Token Index: 271

Text #11

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6113. Min Act: -0.0002

Data Index: 363780 (C4 (Web Text))

Max Activating Token Index: 313

Text #12

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6182. Min Act: -0.0002

Data Index: 1167937 (C4 (Web Text))

Max Activating Token Index: 783

Text #13

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5379. Min Act: -0.0002

Data Index: 784614 (C4 (Web Text))

Max Activating Token Index: 75

Text #14

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5302. Min Act: -0.0002

Data Index: 621606 (C4 (Web Text))

Max Activating Token Index: 815

Text #15

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.6037. Min Act: -0.0002

Data Index: 1071727 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5310. Min Act: -0.0002

Data Index: 1094323 (C4 (Web Text))

Max Activating Token Index: 256

Text #17

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5508. Min Act: -0.0002

Data Index: 486781 (C4 (Web Text))

Max Activating Token Index: 110

Text #18

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5825. Min Act: -0.0002

Data Index: 443494 (C4 (Web Text))

Max Activating Token Index: 188

Text #19

Max Range: 0.6949. Min Range: -0.6949

Max Act: 0.5303. Min Act: -0.0002

Data Index: 1204780 (C4 (Web Text))

Max Activating Token Index: 559