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

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

Neuron 728 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0056. Min Act: -0.0006

Data Index: 389000 (C4 (Web Text))

Max Activating Token Index: 565

Text #1

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0057. Min Act: -0.0006

Data Index: 1050824 (C4 (Web Text))

Max Activating Token Index: 826

Text #2

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0054. Min Act: -0.0006

Data Index: 924037 (C4 (Web Text))

Max Activating Token Index: 152

Text #3

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0051. Min Act: -0.0006

Data Index: 898868 (C4 (Web Text))

Max Activating Token Index: 845

Text #4

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0051. Min Act: -0.0006

Data Index: 1459526 (Python Code)

Max Activating Token Index: 192

Text #5

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0051. Min Act: -0.0006

Data Index: 607414 (C4 (Web Text))

Max Activating Token Index: 175

Text #6

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0049. Min Act: -0.0006

Data Index: 618422 (C4 (Web Text))

Max Activating Token Index: 428

Text #7

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0051. Min Act: -0.0006

Data Index: 293374 (C4 (Web Text))

Max Activating Token Index: 839

Text #8

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0049. Min Act: -0.0006

Data Index: 509822 (C4 (Web Text))

Max Activating Token Index: 186

Text #9

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0049. Min Act: -0.0006

Data Index: 962327 (C4 (Web Text))

Max Activating Token Index: 102

Text #10

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0048. Min Act: -0.0006

Data Index: 103944 (C4 (Web Text))

Max Activating Token Index: 164

Text #11

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0047. Min Act: -0.0006

Data Index: 415010 (C4 (Web Text))

Max Activating Token Index: 239

Text #12

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0048. Min Act: -0.0006

Data Index: 58710 (C4 (Web Text))

Max Activating Token Index: 686

Text #13

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0046. Min Act: -0.0006

Data Index: 348379 (C4 (Web Text))

Max Activating Token Index: 243

Text #14

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0046. Min Act: -0.0006

Data Index: 1368973 (Python Code)

Max Activating Token Index: 611

Text #15

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0047. Min Act: -0.0006

Data Index: 89479 (C4 (Web Text))

Max Activating Token Index: 313

Text #16

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0046. Min Act: -0.0006

Data Index: 517613 (C4 (Web Text))

Max Activating Token Index: 434

Text #17

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0048. Min Act: -0.0006

Data Index: 872592 (C4 (Web Text))

Max Activating Token Index: 303

Text #18

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0048. Min Act: -0.0006

Data Index: 1347126 (C4 (Web Text))

Max Activating Token Index: 751

Text #19

Max Range: 0.0057. Min Range: -0.0057

Max Act: 0.0042. Min Act: -0.0006

Data Index: 827257 (C4 (Web Text))

Max Activating Token Index: 715