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

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

Neuron 1517 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.7885. Min Act: -0.0000

Data Index: 283157 (C4 (Web Text))

Max Activating Token Index: 90

Text #1

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.3817. Min Act: -0.0000

Data Index: 581751 (C4 (Web Text))

Max Activating Token Index: 584

Text #2

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.3174. Min Act: -0.0000

Data Index: 882578 (C4 (Web Text))

Max Activating Token Index: 354

Text #3

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.2388. Min Act: -0.0001

Data Index: 1062203 (C4 (Web Text))

Max Activating Token Index: 251

Text #4

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.3243. Min Act: -0.0001

Data Index: 1230874 (C4 (Web Text))

Max Activating Token Index: 595

Text #5

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.2008. Min Act: -0.0000

Data Index: 796047 (C4 (Web Text))

Max Activating Token Index: 133

Text #6

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.2572. Min Act: -0.0000

Data Index: 727893 (C4 (Web Text))

Max Activating Token Index: 813

Text #7

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.0503. Min Act: -0.0000

Data Index: 848522 (C4 (Web Text))

Max Activating Token Index: 500

Text #8

Max Range: 2.7885. Min Range: -2.7885

Max Act: 2.0934. Min Act: -0.0000

Data Index: 1118498 (C4 (Web Text))

Max Activating Token Index: 149

Text #9

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.9837. Min Act: -0.0000

Data Index: 327081 (C4 (Web Text))

Max Activating Token Index: 1001

Text #10

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.9831. Min Act: -0.0001

Data Index: 548432 (C4 (Web Text))

Max Activating Token Index: 626

Text #11

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.9069. Min Act: -0.0000

Data Index: 54570 (C4 (Web Text))

Max Activating Token Index: 1005

Text #12

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.8694. Min Act: -0.0001

Data Index: 1462964 (Python Code)

Max Activating Token Index: 921

Text #13

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.6076. Min Act: -0.0001

Data Index: 1377107 (Python Code)

Max Activating Token Index: 976

Text #14

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.9410. Min Act: -0.0000

Data Index: 191975 (C4 (Web Text))

Max Activating Token Index: 789

Text #15

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.5291. Min Act: -0.0001

Data Index: 1330818 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.5877. Min Act: -0.0000

Data Index: 1157200 (C4 (Web Text))

Max Activating Token Index: 987

Text #17

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.7123. Min Act: -0.0001

Data Index: 1656130 (Python Code)

Max Activating Token Index: 640

Text #18

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.7488. Min Act: -0.0001

Data Index: 464440 (C4 (Web Text))

Max Activating Token Index: 322

Text #19

Max Range: 2.7885. Min Range: -2.7885

Max Act: 1.5567. Min Act: -0.0000

Data Index: 300524 (C4 (Web Text))

Max Activating Token Index: 312