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

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

Neuron 3985 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1694. Min Act: -0.0000

Data Index: 1525672 (Python Code)

Max Activating Token Index: 849

Text #1

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1924. Min Act: -0.0000

Data Index: 1474530 (Python Code)

Max Activating Token Index: 758

Text #2

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1459. Min Act: -0.0000

Data Index: 687147 (C4 (Web Text))

Max Activating Token Index: 567

Text #3

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1407. Min Act: -0.0000

Data Index: 1430703 (Python Code)

Max Activating Token Index: 785

Text #4

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1272. Min Act: -0.0000

Data Index: 1378262 (Python Code)

Max Activating Token Index: 887

Text #5

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1222. Min Act: -0.0000

Data Index: 1055069 (C4 (Web Text))

Max Activating Token Index: 808

Text #6

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1100. Min Act: -0.0000

Data Index: 873801 (C4 (Web Text))

Max Activating Token Index: 455

Text #7

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1108. Min Act: -0.0000

Data Index: 1510447 (Python Code)

Max Activating Token Index: 938

Text #8

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1088. Min Act: -0.0000

Data Index: 911530 (C4 (Web Text))

Max Activating Token Index: 569

Text #9

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1119. Min Act: -0.0000

Data Index: 799980 (C4 (Web Text))

Max Activating Token Index: 361

Text #10

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1057. Min Act: -0.0000

Data Index: 859576 (C4 (Web Text))

Max Activating Token Index: 507

Text #11

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1029. Min Act: -0.0000

Data Index: 669182 (C4 (Web Text))

Max Activating Token Index: 543

Text #12

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1087. Min Act: -0.0000

Data Index: 1329402 (C4 (Web Text))

Max Activating Token Index: 558

Text #13

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.0876. Min Act: -0.0000

Data Index: 1457260 (Python Code)

Max Activating Token Index: 957

Text #14

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1044. Min Act: -0.0000

Data Index: 669901 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1034. Min Act: -0.0000

Data Index: 1530716 (Python Code)

Max Activating Token Index: 895

Text #16

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1046. Min Act: -0.0000

Data Index: 1369466 (Python Code)

Max Activating Token Index: 1003

Text #17

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1061. Min Act: -0.0000

Data Index: 708637 (C4 (Web Text))

Max Activating Token Index: 961

Text #18

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1002. Min Act: -0.0000

Data Index: 1235224 (C4 (Web Text))

Max Activating Token Index: 519

Text #19

Max Range: 0.1924. Min Range: -0.1924

Max Act: 0.1035. Min Act: -0.0000

Data Index: 1283177 (C4 (Web Text))

Max Activating Token Index: 571