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

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

Neuron 1802 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1414. Min Range: -2.1414

Max Act: 2.1414. Min Act: -0.0000

Data Index: 1001423 (C4 (Web Text))

Max Activating Token Index: 671

Text #1

Max Range: 2.1414. Min Range: -2.1414

Max Act: 2.0141. Min Act: -0.0000

Data Index: 105257 (C4 (Web Text))

Max Activating Token Index: 190

Text #2

Max Range: 2.1414. Min Range: -2.1414

Max Act: 2.1397. Min Act: -0.0000

Data Index: 1137147 (C4 (Web Text))

Max Activating Token Index: 22

Text #3

Max Range: 2.1414. Min Range: -2.1414

Max Act: 2.0679. Min Act: -0.0000

Data Index: 118074 (C4 (Web Text))

Max Activating Token Index: 197

Text #4

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.7408. Min Act: -0.0000

Data Index: 817773 (C4 (Web Text))

Max Activating Token Index: 981

Text #5

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.8512. Min Act: -0.0000

Data Index: 1519419 (Python Code)

Max Activating Token Index: 419

Text #6

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.7391. Min Act: -0.0000

Data Index: 945732 (C4 (Web Text))

Max Activating Token Index: 189

Text #7

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.6878. Min Act: -0.0000

Data Index: 557506 (C4 (Web Text))

Max Activating Token Index: 93

Text #8

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.6036. Min Act: -0.0000

Data Index: 976281 (C4 (Web Text))

Max Activating Token Index: 637

Text #9

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.7080. Min Act: -0.0000

Data Index: 1618395 (Python Code)

Max Activating Token Index: 790

Text #10

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4986. Min Act: -0.0000

Data Index: 1238563 (C4 (Web Text))

Max Activating Token Index: 114

Text #11

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4857. Min Act: -0.0000

Data Index: 1411072 (Python Code)

Max Activating Token Index: 178

Text #12

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4466. Min Act: -0.0000

Data Index: 614539 (C4 (Web Text))

Max Activating Token Index: 394

Text #13

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.6060. Min Act: -0.0000

Data Index: 1490764 (Python Code)

Max Activating Token Index: 462

Text #14

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4937. Min Act: -0.0000

Data Index: 289322 (C4 (Web Text))

Max Activating Token Index: 28

Text #15

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4739. Min Act: -0.0000

Data Index: 497075 (C4 (Web Text))

Max Activating Token Index: 31

Text #16

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.5048. Min Act: -0.0000

Data Index: 1573538 (Python Code)

Max Activating Token Index: 452

Text #17

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4753. Min Act: -0.0000

Data Index: 75429 (C4 (Web Text))

Max Activating Token Index: 824

Text #18

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4179. Min Act: -0.0000

Data Index: 1451902 (Python Code)

Max Activating Token Index: 1022

Text #19

Max Range: 2.1414. Min Range: -2.1414

Max Act: 1.4430. Min Act: -0.0000

Data Index: 252754 (C4 (Web Text))

Max Activating Token Index: 948