Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1980 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1201. Min Act: -0.0001

Data Index: 1519703 (Python Code)

Max Activating Token Index: 991

Text #1

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1338. Min Act: -0.0001

Data Index: 828938 (C4 (Web Text))

Max Activating Token Index: 155

Text #2

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1074. Min Act: -0.0000

Data Index: 1203497 (C4 (Web Text))

Max Activating Token Index: 446

Text #3

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1074. Min Act: -0.0001

Data Index: 1591579 (Python Code)

Max Activating Token Index: 982

Text #4

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0897. Min Act: -0.0000

Data Index: 247397 (C4 (Web Text))

Max Activating Token Index: 1020

Text #5

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1161. Min Act: -0.0000

Data Index: 575800 (C4 (Web Text))

Max Activating Token Index: 169

Text #6

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1079. Min Act: -0.0001

Data Index: 528226 (C4 (Web Text))

Max Activating Token Index: 324

Text #7

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0991. Min Act: -0.0000

Data Index: 136789 (C4 (Web Text))

Max Activating Token Index: 110

Text #8

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0852. Min Act: -0.0001

Data Index: 1219396 (C4 (Web Text))

Max Activating Token Index: 196

Text #9

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0938. Min Act: -0.0001

Data Index: 8667 (C4 (Web Text))

Max Activating Token Index: 176

Text #10

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0943. Min Act: -0.0001

Data Index: 1643637 (Python Code)

Max Activating Token Index: 829

Text #11

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0863. Min Act: -0.0000

Data Index: 401544 (C4 (Web Text))

Max Activating Token Index: 411

Text #12

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.1001. Min Act: -0.0000

Data Index: 63207 (C4 (Web Text))

Max Activating Token Index: 968

Text #13

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0804. Min Act: -0.0001

Data Index: 1641942 (Python Code)

Max Activating Token Index: 728

Text #14

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0837. Min Act: -0.0000

Data Index: 6101 (C4 (Web Text))

Max Activating Token Index: 1012

Text #15

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0919. Min Act: -0.0001

Data Index: 1345129 (C4 (Web Text))

Max Activating Token Index: 224

Text #16

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0878. Min Act: -0.0000

Data Index: 597443 (C4 (Web Text))

Max Activating Token Index: 963

Text #17

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0810. Min Act: -0.0000

Data Index: 1366055 (Python Code)

Max Activating Token Index: 729

Text #18

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0880. Min Act: -0.0001

Data Index: 1036297 (C4 (Web Text))

Max Activating Token Index: 337

Text #19

Max Range: 0.1338. Min Range: -0.1338

Max Act: 0.0826. Min Act: -0.0000

Data Index: 578793 (C4 (Web Text))

Max Activating Token Index: 976