Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 19 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1227. Min Act: -0.0002

Data Index: 829572 (C4 (Web Text))

Max Activating Token Index: 365

Text #1

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1222. Min Act: -0.0002

Data Index: 606623 (C4 (Web Text))

Max Activating Token Index: 720

Text #2

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1129. Min Act: -0.0002

Data Index: 1550207 (Python Code)

Max Activating Token Index: 126

Text #3

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1120. Min Act: -0.0002

Data Index: 1027422 (C4 (Web Text))

Max Activating Token Index: 766

Text #4

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1169. Min Act: -0.0002

Data Index: 262148 (C4 (Web Text))

Max Activating Token Index: 684

Text #5

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1080. Min Act: -0.0002

Data Index: 636886 (C4 (Web Text))

Max Activating Token Index: 668

Text #6

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1080. Min Act: -0.0002

Data Index: 1505620 (Python Code)

Max Activating Token Index: 109

Text #7

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1037. Min Act: -0.0002

Data Index: 1633046 (Python Code)

Max Activating Token Index: 463

Text #8

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1050. Min Act: -0.0002

Data Index: 970117 (C4 (Web Text))

Max Activating Token Index: 725

Text #9

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1042. Min Act: -0.0002

Data Index: 109693 (C4 (Web Text))

Max Activating Token Index: 618

Text #10

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1049. Min Act: -0.0002

Data Index: 910025 (C4 (Web Text))

Max Activating Token Index: 795

Text #11

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1046. Min Act: -0.0002

Data Index: 1035332 (C4 (Web Text))

Max Activating Token Index: 953

Text #12

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1060. Min Act: -0.0002

Data Index: 1528090 (Python Code)

Max Activating Token Index: 848

Text #13

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1035. Min Act: -0.0002

Data Index: 95881 (C4 (Web Text))

Max Activating Token Index: 776

Text #14

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.0982. Min Act: -0.0002

Data Index: 142598 (C4 (Web Text))

Max Activating Token Index: 906

Text #15

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1042. Min Act: -0.0002

Data Index: 497681 (C4 (Web Text))

Max Activating Token Index: 803

Text #16

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1028. Min Act: -0.0002

Data Index: 1577537 (Python Code)

Max Activating Token Index: 510

Text #17

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1020. Min Act: -0.0002

Data Index: 1441457 (Python Code)

Max Activating Token Index: 17

Text #18

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.0988. Min Act: -0.0002

Data Index: 1055127 (C4 (Web Text))

Max Activating Token Index: 102

Text #19

Max Range: 0.1227. Min Range: -0.1227

Max Act: 0.1039. Min Act: -0.0002

Data Index: 121007 (C4 (Web Text))

Max Activating Token Index: 268