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

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

Neuron 32 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: 0.4506. Min Range: -0.4506

Max Act: 0.4506. Min Act: -0.0001

Data Index: 922624 (C4 (Web Text))

Max Activating Token Index: 749

Text #1

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.4082. Min Act: -0.0001

Data Index: 1148262 (C4 (Web Text))

Max Activating Token Index: 36

Text #2

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3965. Min Act: -0.0001

Data Index: 564440 (C4 (Web Text))

Max Activating Token Index: 833

Text #3

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3869. Min Act: -0.0001

Data Index: 807794 (C4 (Web Text))

Max Activating Token Index: 780

Text #4

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3798. Min Act: -0.0001

Data Index: 637713 (C4 (Web Text))

Max Activating Token Index: 159

Text #5

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3620. Min Act: -0.0001

Data Index: 843699 (C4 (Web Text))

Max Activating Token Index: 102

Text #6

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3807. Min Act: -0.0001

Data Index: 570925 (C4 (Web Text))

Max Activating Token Index: 131

Text #7

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3542. Min Act: -0.0001

Data Index: 474157 (C4 (Web Text))

Max Activating Token Index: 709

Text #8

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3609. Min Act: -0.0001

Data Index: 1060098 (C4 (Web Text))

Max Activating Token Index: 528

Text #9

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3249. Min Act: -0.0001

Data Index: 1121603 (C4 (Web Text))

Max Activating Token Index: 867

Text #10

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3347. Min Act: -0.0001

Data Index: 1097714 (C4 (Web Text))

Max Activating Token Index: 688

Text #11

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3541. Min Act: -0.0001

Data Index: 731022 (C4 (Web Text))

Max Activating Token Index: 751

Text #12

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3196. Min Act: -0.0000

Data Index: 1136657 (C4 (Web Text))

Max Activating Token Index: 38

Text #13

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3122. Min Act: -0.0001

Data Index: 1646495 (Python Code)

Max Activating Token Index: 223

Text #14

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3202. Min Act: -0.0001

Data Index: 1513269 (Python Code)

Max Activating Token Index: 189

Text #15

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3158. Min Act: -0.0001

Data Index: 751632 (C4 (Web Text))

Max Activating Token Index: 793

Text #16

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.2933. Min Act: -0.0001

Data Index: 1134190 (C4 (Web Text))

Max Activating Token Index: 703

Text #17

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.2999. Min Act: -0.0001

Data Index: 170010 (C4 (Web Text))

Max Activating Token Index: 758

Text #18

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.2980. Min Act: -0.0001

Data Index: 1446918 (Python Code)

Max Activating Token Index: 382

Text #19

Max Range: 0.4506. Min Range: -0.4506

Max Act: 0.3092. Min Act: 0.0000

Data Index: 1008159 (C4 (Web Text))

Max Activating Token Index: 926