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

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

Neuron 1631 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: 3.7482. Min Range: -3.7482

Max Act: 3.7482. Min Act: -0.0001

Data Index: 1538322 (Python Code)

Max Activating Token Index: 929

Text #1

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.5718. Min Act: -0.0001

Data Index: 235133 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.3420. Min Act: -0.0001

Data Index: 272202 (C4 (Web Text))

Max Activating Token Index: 597

Text #3

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.5905. Min Act: -0.0001

Data Index: 807059 (C4 (Web Text))

Max Activating Token Index: 930

Text #4

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0568. Min Act: -0.0001

Data Index: 269519 (C4 (Web Text))

Max Activating Token Index: 75

Text #5

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.1893. Min Act: -0.0001

Data Index: 1370425 (Python Code)

Max Activating Token Index: 420

Text #6

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0505. Min Act: -0.0001

Data Index: 1594770 (Python Code)

Max Activating Token Index: 181

Text #7

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.9083. Min Act: -0.0001

Data Index: 69092 (C4 (Web Text))

Max Activating Token Index: 328

Text #8

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.9841. Min Act: -0.0001

Data Index: 1448822 (Python Code)

Max Activating Token Index: 95

Text #9

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.1238. Min Act: -0.0001

Data Index: 435820 (C4 (Web Text))

Max Activating Token Index: 723

Text #10

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.9815. Min Act: -0.0001

Data Index: 1223259 (C4 (Web Text))

Max Activating Token Index: 670

Text #11

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.9400. Min Act: -0.0001

Data Index: 1368979 (Python Code)

Max Activating Token Index: 44

Text #12

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.9383. Min Act: -0.0001

Data Index: 1625770 (Python Code)

Max Activating Token Index: 37

Text #13

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.8004. Min Act: -0.0001

Data Index: 1349302 (C4 (Web Text))

Max Activating Token Index: 369

Text #14

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0407. Min Act: -0.0001

Data Index: 150302 (C4 (Web Text))

Max Activating Token Index: 292

Text #15

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.7910. Min Act: -0.0001

Data Index: 346488 (C4 (Web Text))

Max Activating Token Index: 256

Text #16

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0429. Min Act: -0.0001

Data Index: 1567718 (Python Code)

Max Activating Token Index: 371

Text #17

Max Range: 3.7482. Min Range: -3.7482

Max Act: 2.8869. Min Act: -0.0001

Data Index: 1466687 (Python Code)

Max Activating Token Index: 4

Text #18

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0167. Min Act: -0.0001

Data Index: 809357 (C4 (Web Text))

Max Activating Token Index: 356

Text #19

Max Range: 3.7482. Min Range: -3.7482

Max Act: 3.0193. Min Act: -0.0001

Data Index: 242698 (C4 (Web Text))

Max Activating Token Index: 119