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

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

Neuron 895 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1538. Min Act: -0.0000

Data Index: 590656 (C4 (Web Text))

Max Activating Token Index: 829

Text #1

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1199. Min Act: -0.0000

Data Index: 792232 (C4 (Web Text))

Max Activating Token Index: 586

Text #2

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1163. Min Act: -0.0000

Data Index: 895596 (C4 (Web Text))

Max Activating Token Index: 970

Text #3

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1111. Min Act: -0.0000

Data Index: 1610187 (Python Code)

Max Activating Token Index: 484

Text #4

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1191. Min Act: -0.0000

Data Index: 46499 (C4 (Web Text))

Max Activating Token Index: 976

Text #5

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1083. Min Act: -0.0000

Data Index: 396055 (C4 (Web Text))

Max Activating Token Index: 970

Text #6

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1059. Min Act: -0.0000

Data Index: 1254419 (C4 (Web Text))

Max Activating Token Index: 354

Text #7

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1009. Min Act: -0.0000

Data Index: 1589726 (Python Code)

Max Activating Token Index: 535

Text #8

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1068. Min Act: -0.0000

Data Index: 1065425 (C4 (Web Text))

Max Activating Token Index: 761

Text #9

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0992. Min Act: -0.0000

Data Index: 560330 (C4 (Web Text))

Max Activating Token Index: 266

Text #10

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0913. Min Act: -0.0000

Data Index: 745312 (C4 (Web Text))

Max Activating Token Index: 889

Text #11

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0932. Min Act: -0.0000

Data Index: 1634964 (Python Code)

Max Activating Token Index: 377

Text #12

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.1018. Min Act: -0.0000

Data Index: 1599085 (Python Code)

Max Activating Token Index: 825

Text #13

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0972. Min Act: -0.0000

Data Index: 1475097 (Python Code)

Max Activating Token Index: 906

Text #14

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0936. Min Act: -0.0000

Data Index: 1468276 (Python Code)

Max Activating Token Index: 532

Text #15

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0900. Min Act: -0.0000

Data Index: 674600 (C4 (Web Text))

Max Activating Token Index: 421

Text #16

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0865. Min Act: -0.0000

Data Index: 1220199 (C4 (Web Text))

Max Activating Token Index: 951

Text #17

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0997. Min Act: -0.0000

Data Index: 1191124 (C4 (Web Text))

Max Activating Token Index: 927

Text #18

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0883. Min Act: -0.0000

Data Index: 73556 (C4 (Web Text))

Max Activating Token Index: 956

Text #19

Max Range: 0.1538. Min Range: -0.1538

Max Act: 0.0781. Min Act: -0.0000

Data Index: 1122691 (C4 (Web Text))

Max Activating Token Index: 878