Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1113 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0058. Min Act: -0.0003

Data Index: 385344 (C4 (Web Text))

Max Activating Token Index: 291

Text #1

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0047. Min Act: -0.0003

Data Index: 949579 (C4 (Web Text))

Max Activating Token Index: 86

Text #2

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0040. Min Act: -0.0002

Data Index: 454472 (C4 (Web Text))

Max Activating Token Index: 571

Text #3

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0040. Min Act: -0.0003

Data Index: 1548640 (Python Code)

Max Activating Token Index: 633

Text #4

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0038. Min Act: -0.0003

Data Index: 1635946 (Python Code)

Max Activating Token Index: 189

Text #5

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0035. Min Act: -0.0003

Data Index: 860577 (C4 (Web Text))

Max Activating Token Index: 658

Text #6

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0033. Min Act: -0.0004

Data Index: 474177 (C4 (Web Text))

Max Activating Token Index: 59

Text #7

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0034. Min Act: -0.0003

Data Index: 878946 (C4 (Web Text))

Max Activating Token Index: 757

Text #8

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0029. Min Act: -0.0003

Data Index: 1510470 (Python Code)

Max Activating Token Index: 785

Text #9

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0028. Min Act: -0.0003

Data Index: 1460062 (Python Code)

Max Activating Token Index: 476

Text #10

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0030. Min Act: -0.0003

Data Index: 185365 (C4 (Web Text))

Max Activating Token Index: 388

Text #11

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0027. Min Act: -0.0003

Data Index: 1379081 (Python Code)

Max Activating Token Index: 777

Text #12

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0028. Min Act: -0.0003

Data Index: 1462907 (Python Code)

Max Activating Token Index: 885

Text #13

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0030. Min Act: -0.0003

Data Index: 1188168 (C4 (Web Text))

Max Activating Token Index: 576

Text #14

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0026. Min Act: -0.0003

Data Index: 1109980 (C4 (Web Text))

Max Activating Token Index: 508

Text #15

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0027. Min Act: -0.0003

Data Index: 595633 (C4 (Web Text))

Max Activating Token Index: 219

Text #16

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0028. Min Act: -0.0003

Data Index: 1591940 (Python Code)

Max Activating Token Index: 986

Text #17

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0028. Min Act: -0.0003

Data Index: 1585413 (Python Code)

Max Activating Token Index: 964

Text #18

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0026. Min Act: -0.0003

Data Index: 1533698 (Python Code)

Max Activating Token Index: 28

Text #19

Max Range: 0.0058. Min Range: -0.0058

Max Act: 0.0024. Min Act: -0.0003

Data Index: 59308 (C4 (Web Text))

Max Activating Token Index: 52