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

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

Neuron 923 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.4385. Min Act: -0.0001

Data Index: 970063 (C4 (Web Text))

Max Activating Token Index: 221

Text #1

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.3551. Min Act: -0.0001

Data Index: 177048 (C4 (Web Text))

Max Activating Token Index: 400

Text #2

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.3488. Min Act: -0.0001

Data Index: 878028 (C4 (Web Text))

Max Activating Token Index: 382

Text #3

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2526. Min Act: -0.0001

Data Index: 1117940 (C4 (Web Text))

Max Activating Token Index: 918

Text #4

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2839. Min Act: -0.0001

Data Index: 795759 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2656. Min Act: -0.0001

Data Index: 809891 (C4 (Web Text))

Max Activating Token Index: 792

Text #6

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2545. Min Act: -0.0001

Data Index: 28935 (C4 (Web Text))

Max Activating Token Index: 264

Text #7

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2807. Min Act: -0.0001

Data Index: 336192 (C4 (Web Text))

Max Activating Token Index: 154

Text #8

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.3162. Min Act: -0.0001

Data Index: 1043138 (C4 (Web Text))

Max Activating Token Index: 366

Text #9

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2098. Min Act: -0.0001

Data Index: 1103251 (C4 (Web Text))

Max Activating Token Index: 652

Text #10

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.2827. Min Act: -0.0001

Data Index: 1349272 (C4 (Web Text))

Max Activating Token Index: 606

Text #11

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.1137. Min Act: -0.0001

Data Index: 1525674 (Python Code)

Max Activating Token Index: 950

Text #12

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.1239. Min Act: -0.0001

Data Index: 680666 (C4 (Web Text))

Max Activating Token Index: 817

Text #13

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.0597. Min Act: -0.0001

Data Index: 324210 (C4 (Web Text))

Max Activating Token Index: 274

Text #14

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.1277. Min Act: -0.0001

Data Index: 30441 (C4 (Web Text))

Max Activating Token Index: 133

Text #15

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.0751. Min Act: -0.0001

Data Index: 98524 (C4 (Web Text))

Max Activating Token Index: 212

Text #16

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.0912. Min Act: -0.0001

Data Index: 1507763 (Python Code)

Max Activating Token Index: 933

Text #17

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.0543. Min Act: -0.0001

Data Index: 107549 (C4 (Web Text))

Max Activating Token Index: 508

Text #18

Max Range: 1.4385. Min Range: -1.4385

Max Act: 0.9899. Min Act: -0.0001

Data Index: 214713 (C4 (Web Text))

Max Activating Token Index: 500

Text #19

Max Range: 1.4385. Min Range: -1.4385

Max Act: 1.0731. Min Act: -0.0001

Data Index: 8020 (C4 (Web Text))

Max Activating Token Index: 913