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

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

Neuron 3126 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2716. Min Act: -0.0000

Data Index: 1481398 (Python Code)

Max Activating Token Index: 60

Text #1

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2668. Min Act: -0.0001

Data Index: 86132 (C4 (Web Text))

Max Activating Token Index: 633

Text #2

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2528. Min Act: -0.0001

Data Index: 243617 (C4 (Web Text))

Max Activating Token Index: 820

Text #3

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2354. Min Act: -0.0000

Data Index: 1336963 (C4 (Web Text))

Max Activating Token Index: 730

Text #4

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2384. Min Act: -0.0001

Data Index: 416839 (C4 (Web Text))

Max Activating Token Index: 346

Text #5

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2382. Min Act: -0.0001

Data Index: 410905 (C4 (Web Text))

Max Activating Token Index: 74

Text #6

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2142. Min Act: -0.0001

Data Index: 1314908 (C4 (Web Text))

Max Activating Token Index: 308

Text #7

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2276. Min Act: -0.0001

Data Index: 171273 (C4 (Web Text))

Max Activating Token Index: 343

Text #8

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2418. Min Act: -0.0000

Data Index: 1561692 (Python Code)

Max Activating Token Index: 739

Text #9

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2028. Min Act: -0.0001

Data Index: 841337 (C4 (Web Text))

Max Activating Token Index: 845

Text #10

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2123. Min Act: -0.0001

Data Index: 941498 (C4 (Web Text))

Max Activating Token Index: 50

Text #11

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2041. Min Act: -0.0001

Data Index: 730748 (C4 (Web Text))

Max Activating Token Index: 952

Text #12

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2088. Min Act: -0.0001

Data Index: 1261868 (C4 (Web Text))

Max Activating Token Index: 524

Text #13

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2030. Min Act: -0.0001

Data Index: 974126 (C4 (Web Text))

Max Activating Token Index: 707

Text #14

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2083. Min Act: -0.0001

Data Index: 621630 (C4 (Web Text))

Max Activating Token Index: 955

Text #15

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2042. Min Act: -0.0001

Data Index: 373068 (C4 (Web Text))

Max Activating Token Index: 667

Text #16

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.2008. Min Act: -0.0000

Data Index: 1620505 (Python Code)

Max Activating Token Index: 164

Text #17

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.1952. Min Act: -0.0001

Data Index: 1289033 (C4 (Web Text))

Max Activating Token Index: 352

Text #18

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.1908. Min Act: -0.0001

Data Index: 336361 (C4 (Web Text))

Max Activating Token Index: 895

Text #19

Max Range: 0.2716. Min Range: -0.2716

Max Act: 0.1949. Min Act: -0.0001

Data Index: 307715 (C4 (Web Text))

Max Activating Token Index: 20