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

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

Neuron 2445 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1678. Min Range: -2.1678

Max Act: 2.1082. Min Act: -0.0000

Data Index: 358287 (C4 (Web Text))

Max Activating Token Index: 106

Text #1

Max Range: 2.1678. Min Range: -2.1678

Max Act: 2.1129. Min Act: -0.0000

Data Index: 1051774 (C4 (Web Text))

Max Activating Token Index: 919

Text #2

Max Range: 2.1678. Min Range: -2.1678

Max Act: 2.1678. Min Act: -0.0000

Data Index: 424921 (C4 (Web Text))

Max Activating Token Index: 420

Text #3

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.7287. Min Act: -0.0000

Data Index: 208217 (C4 (Web Text))

Max Activating Token Index: 380

Text #4

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.7935. Min Act: -0.0000

Data Index: 1521439 (Python Code)

Max Activating Token Index: 876

Text #5

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.6991. Min Act: -0.0000

Data Index: 1399199 (Python Code)

Max Activating Token Index: 153

Text #6

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5608. Min Act: -0.0000

Data Index: 365271 (C4 (Web Text))

Max Activating Token Index: 408

Text #7

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5259. Min Act: -0.0000

Data Index: 1396221 (Python Code)

Max Activating Token Index: 355

Text #8

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.6371. Min Act: -0.0000

Data Index: 533764 (C4 (Web Text))

Max Activating Token Index: 821

Text #9

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5773. Min Act: -0.0000

Data Index: 467503 (C4 (Web Text))

Max Activating Token Index: 954

Text #10

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5796. Min Act: -0.0000

Data Index: 334551 (C4 (Web Text))

Max Activating Token Index: 590

Text #11

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5794. Min Act: -0.0000

Data Index: 202198 (C4 (Web Text))

Max Activating Token Index: 857

Text #12

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5512. Min Act: -0.0000

Data Index: 1624303 (Python Code)

Max Activating Token Index: 908

Text #13

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.4562. Min Act: -0.0000

Data Index: 963693 (C4 (Web Text))

Max Activating Token Index: 638

Text #14

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.5366. Min Act: -0.0000

Data Index: 828029 (C4 (Web Text))

Max Activating Token Index: 85

Text #15

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.4199. Min Act: -0.0000

Data Index: 694964 (C4 (Web Text))

Max Activating Token Index: 164

Text #16

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.6026. Min Act: -0.0000

Data Index: 1367024 (Python Code)

Max Activating Token Index: 421

Text #17

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.4784. Min Act: -0.0000

Data Index: 987991 (C4 (Web Text))

Max Activating Token Index: 181

Text #18

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.4005. Min Act: -0.0000

Data Index: 916600 (C4 (Web Text))

Max Activating Token Index: 861

Text #19

Max Range: 2.1678. Min Range: -2.1678

Max Act: 1.4393. Min Act: -0.0000

Data Index: 139003 (C4 (Web Text))

Max Activating Token Index: 935