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

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

Neuron 3052 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.7852. Min Act: -0.0000

Data Index: 418911 (C4 (Web Text))

Max Activating Token Index: 41

Text #1

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.6595. Min Act: -0.0000

Data Index: 710948 (C4 (Web Text))

Max Activating Token Index: 46

Text #2

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.2195. Min Act: -0.0000

Data Index: 1578241 (Python Code)

Max Activating Token Index: 632

Text #3

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.2390. Min Act: -0.0000

Data Index: 1070173 (C4 (Web Text))

Max Activating Token Index: 109

Text #4

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.1450. Min Act: -0.0000

Data Index: 792544 (C4 (Web Text))

Max Activating Token Index: 425

Text #5

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.0650. Min Act: -0.0000

Data Index: 132120 (C4 (Web Text))

Max Activating Token Index: 180

Text #6

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.1351. Min Act: -0.0000

Data Index: 278356 (C4 (Web Text))

Max Activating Token Index: 253

Text #7

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.0678. Min Act: -0.0000

Data Index: 641555 (C4 (Web Text))

Max Activating Token Index: 132

Text #8

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.9878. Min Act: -0.0000

Data Index: 743695 (C4 (Web Text))

Max Activating Token Index: 58

Text #9

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.9833. Min Act: -0.0000

Data Index: 1323066 (C4 (Web Text))

Max Activating Token Index: 604

Text #10

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.9246. Min Act: -0.0000

Data Index: 379110 (C4 (Web Text))

Max Activating Token Index: 787

Text #11

Max Range: 1.7852. Min Range: -1.7852

Max Act: 1.0062. Min Act: -0.0000

Data Index: 1374985 (Python Code)

Max Activating Token Index: 404

Text #12

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.9036. Min Act: -0.0000

Data Index: 1347070 (C4 (Web Text))

Max Activating Token Index: 95

Text #13

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8943. Min Act: -0.0000

Data Index: 1402741 (Python Code)

Max Activating Token Index: 712

Text #14

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.9178. Min Act: -0.0000

Data Index: 199748 (C4 (Web Text))

Max Activating Token Index: 171

Text #15

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8639. Min Act: -0.0000

Data Index: 1525315 (Python Code)

Max Activating Token Index: 109

Text #16

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8834. Min Act: -0.0000

Data Index: 1383811 (Python Code)

Max Activating Token Index: 408

Text #17

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8362. Min Act: -0.0000

Data Index: 1516931 (Python Code)

Max Activating Token Index: 379

Text #18

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8885. Min Act: -0.0000

Data Index: 857175 (C4 (Web Text))

Max Activating Token Index: 160

Text #19

Max Range: 1.7852. Min Range: -1.7852

Max Act: 0.8309. Min Act: -0.0000

Data Index: 1372000 (Python Code)

Max Activating Token Index: 460