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

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

Neuron 4405 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1945. Min Range: -1.1945

Max Act: 1.1945. Min Act: -0.0000

Data Index: 454375 (C4 (Web Text))

Max Activating Token Index: 777

Text #1

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.9903. Min Act: -0.0000

Data Index: 661891 (C4 (Web Text))

Max Activating Token Index: 756

Text #2

Max Range: 1.1945. Min Range: -1.1945

Max Act: 1.0516. Min Act: -0.0000

Data Index: 96075 (C4 (Web Text))

Max Activating Token Index: 871

Text #3

Max Range: 1.1945. Min Range: -1.1945

Max Act: 1.0071. Min Act: -0.0000

Data Index: 335013 (C4 (Web Text))

Max Activating Token Index: 805

Text #4

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.9516. Min Act: -0.0000

Data Index: 540533 (C4 (Web Text))

Max Activating Token Index: 921

Text #5

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.9103. Min Act: -0.0000

Data Index: 347762 (C4 (Web Text))

Max Activating Token Index: 386

Text #6

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8167. Min Act: -0.0000

Data Index: 62051 (C4 (Web Text))

Max Activating Token Index: 845

Text #7

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7619. Min Act: -0.0000

Data Index: 114730 (C4 (Web Text))

Max Activating Token Index: 244

Text #8

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8149. Min Act: -0.0000

Data Index: 665744 (C4 (Web Text))

Max Activating Token Index: 709

Text #9

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8835. Min Act: -0.0000

Data Index: 947410 (C4 (Web Text))

Max Activating Token Index: 796

Text #10

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8113. Min Act: -0.0001

Data Index: 930312 (C4 (Web Text))

Max Activating Token Index: 210

Text #11

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8107. Min Act: -0.0000

Data Index: 621777 (C4 (Web Text))

Max Activating Token Index: 154

Text #12

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8053. Min Act: -0.0000

Data Index: 1230340 (C4 (Web Text))

Max Activating Token Index: 977

Text #13

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7993. Min Act: -0.0000

Data Index: 901812 (C4 (Web Text))

Max Activating Token Index: 848

Text #14

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7441. Min Act: -0.0000

Data Index: 501045 (C4 (Web Text))

Max Activating Token Index: 231

Text #15

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7898. Min Act: -0.0000

Data Index: 402214 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7043. Min Act: -0.0000

Data Index: 1501849 (Python Code)

Max Activating Token Index: 725

Text #17

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.8472. Min Act: -0.0000

Data Index: 406604 (C4 (Web Text))

Max Activating Token Index: 41

Text #18

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7990. Min Act: -0.0000

Data Index: 413946 (C4 (Web Text))

Max Activating Token Index: 150

Text #19

Max Range: 1.1945. Min Range: -1.1945

Max Act: 0.7952. Min Act: -0.0000

Data Index: 759060 (C4 (Web Text))

Max Activating Token Index: 272