Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4529 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1465. Min Act: -0.0000

Data Index: 620025 (C4 (Web Text))

Max Activating Token Index: 505

Text #1

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1302. Min Act: -0.0000

Data Index: 265837 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1433. Min Act: -0.0000

Data Index: 74883 (C4 (Web Text))

Max Activating Token Index: 467

Text #3

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1415. Min Act: -0.0000

Data Index: 1154652 (C4 (Web Text))

Max Activating Token Index: 139

Text #4

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1005. Min Act: -0.0000

Data Index: 1569610 (Python Code)

Max Activating Token Index: 958

Text #5

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1283. Min Act: -0.0000

Data Index: 259444 (C4 (Web Text))

Max Activating Token Index: 490

Text #6

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.0935. Min Act: -0.0000

Data Index: 70763 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.0966. Min Act: -0.0000

Data Index: 1312078 (C4 (Web Text))

Max Activating Token Index: 316

Text #8

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1056. Min Act: -0.0000

Data Index: 1381800 (Python Code)

Max Activating Token Index: 62

Text #9

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1026. Min Act: -0.0000

Data Index: 1597715 (Python Code)

Max Activating Token Index: 294

Text #10

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1043. Min Act: -0.0000

Data Index: 1578102 (Python Code)

Max Activating Token Index: 159

Text #11

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1018. Min Act: -0.0000

Data Index: 1220449 (C4 (Web Text))

Max Activating Token Index: 479

Text #12

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1070. Min Act: -0.0000

Data Index: 1588960 (Python Code)

Max Activating Token Index: 265

Text #13

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1088. Min Act: -0.0000

Data Index: 42226 (C4 (Web Text))

Max Activating Token Index: 8

Text #14

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1200. Min Act: -0.0000

Data Index: 296948 (C4 (Web Text))

Max Activating Token Index: 235

Text #15

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1045. Min Act: -0.0000

Data Index: 1518234 (Python Code)

Max Activating Token Index: 170

Text #16

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.0876. Min Act: -0.0000

Data Index: 1007033 (C4 (Web Text))

Max Activating Token Index: 329

Text #17

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1070. Min Act: -0.0000

Data Index: 1193239 (C4 (Web Text))

Max Activating Token Index: 14

Text #18

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.1150. Min Act: -0.0000

Data Index: 1615476 (Python Code)

Max Activating Token Index: 936

Text #19

Max Range: 0.1465. Min Range: -0.1465

Max Act: 0.0899. Min Act: -0.0000

Data Index: 1134750 (C4 (Web Text))

Max Activating Token Index: 716