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

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

Neuron 1087 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2520. Min Act: -0.0000

Data Index: 332742 (C4 (Web Text))

Max Activating Token Index: 375

Text #1

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2289. Min Act: -0.0000

Data Index: 314307 (C4 (Web Text))

Max Activating Token Index: 849

Text #2

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2208. Min Act: -0.0000

Data Index: 1008564 (C4 (Web Text))

Max Activating Token Index: 460

Text #3

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2108. Min Act: -0.0000

Data Index: 351865 (C4 (Web Text))

Max Activating Token Index: 502

Text #4

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2184. Min Act: -0.0000

Data Index: 439688 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2261. Min Act: -0.0000

Data Index: 1126589 (C4 (Web Text))

Max Activating Token Index: 868

Text #6

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2206. Min Act: -0.0000

Data Index: 861557 (C4 (Web Text))

Max Activating Token Index: 767

Text #7

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2103. Min Act: -0.0000

Data Index: 1340103 (C4 (Web Text))

Max Activating Token Index: 200

Text #8

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2107. Min Act: -0.0000

Data Index: 1311099 (C4 (Web Text))

Max Activating Token Index: 968

Text #9

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2125. Min Act: -0.0000

Data Index: 1405201 (Python Code)

Max Activating Token Index: 467

Text #10

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2146. Min Act: -0.0000

Data Index: 324257 (C4 (Web Text))

Max Activating Token Index: 825

Text #11

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1901. Min Act: -0.0000

Data Index: 1432945 (Python Code)

Max Activating Token Index: 734

Text #12

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2014. Min Act: -0.0000

Data Index: 876148 (C4 (Web Text))

Max Activating Token Index: 735

Text #13

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.2128. Min Act: -0.0000

Data Index: 1462045 (Python Code)

Max Activating Token Index: 562

Text #14

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1939. Min Act: -0.0000

Data Index: 679037 (C4 (Web Text))

Max Activating Token Index: 301

Text #15

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1958. Min Act: -0.0000

Data Index: 1303880 (C4 (Web Text))

Max Activating Token Index: 992

Text #16

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1864. Min Act: -0.0000

Data Index: 1047674 (C4 (Web Text))

Max Activating Token Index: 260

Text #17

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1911. Min Act: -0.0000

Data Index: 1073988 (C4 (Web Text))

Max Activating Token Index: 453

Text #18

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1837. Min Act: -0.0000

Data Index: 789404 (C4 (Web Text))

Max Activating Token Index: 276

Text #19

Max Range: 0.2520. Min Range: -0.2520

Max Act: 0.1903. Min Act: -0.0000

Data Index: 137708 (C4 (Web Text))

Max Activating Token Index: 604