Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2017 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5527. Min Act: -0.0001

Data Index: 203796 (C4 (Web Text))

Max Activating Token Index: 201

Text #1

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5867. Min Act: -0.0000

Data Index: 1211960 (C4 (Web Text))

Max Activating Token Index: 267

Text #2

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5455. Min Act: -0.0001

Data Index: 1288324 (C4 (Web Text))

Max Activating Token Index: 319

Text #3

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5228. Min Act: -0.0001

Data Index: 800687 (C4 (Web Text))

Max Activating Token Index: 759

Text #4

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4954. Min Act: -0.0001

Data Index: 339206 (C4 (Web Text))

Max Activating Token Index: 517

Text #5

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5055. Min Act: -0.0000

Data Index: 774856 (C4 (Web Text))

Max Activating Token Index: 232

Text #6

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4864. Min Act: -0.0000

Data Index: 1041469 (C4 (Web Text))

Max Activating Token Index: 177

Text #7

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4728. Min Act: -0.0000

Data Index: 1069305 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.5112. Min Act: -0.0000

Data Index: 210478 (C4 (Web Text))

Max Activating Token Index: 837

Text #9

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4791. Min Act: -0.0000

Data Index: 1156368 (C4 (Web Text))

Max Activating Token Index: 239

Text #10

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4761. Min Act: -0.0000

Data Index: 361652 (C4 (Web Text))

Max Activating Token Index: 125

Text #11

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4554. Min Act: -0.0001

Data Index: 37672 (C4 (Web Text))

Max Activating Token Index: 904

Text #12

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4662. Min Act: -0.0000

Data Index: 179584 (C4 (Web Text))

Max Activating Token Index: 658

Text #13

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4815. Min Act: -0.0000

Data Index: 109495 (C4 (Web Text))

Max Activating Token Index: 273

Text #14

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4656. Min Act: -0.0000

Data Index: 280655 (C4 (Web Text))

Max Activating Token Index: 324

Text #15

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4433. Min Act: -0.0000

Data Index: 139991 (C4 (Web Text))

Max Activating Token Index: 446

Text #16

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4421. Min Act: -0.0000

Data Index: 269084 (C4 (Web Text))

Max Activating Token Index: 524

Text #17

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4459. Min Act: -0.0001

Data Index: 1034409 (C4 (Web Text))

Max Activating Token Index: 25

Text #18

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4430. Min Act: -0.0001

Data Index: 1318431 (C4 (Web Text))

Max Activating Token Index: 100

Text #19

Max Range: 0.5867. Min Range: -0.5867

Max Act: 0.4438. Min Act: -0.0000

Data Index: 886372 (C4 (Web Text))

Max Activating Token Index: 123