Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3487 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.7981. Min Act: -0.0001

Data Index: 223525 (C4 (Web Text))

Max Activating Token Index: 18

Text #1

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.8226. Min Act: -0.0001

Data Index: 465024 (C4 (Web Text))

Max Activating Token Index: 805

Text #2

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.7482. Min Act: -0.0001

Data Index: 724983 (C4 (Web Text))

Max Activating Token Index: 121

Text #3

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.7368. Min Act: -0.0001

Data Index: 1581813 (Python Code)

Max Activating Token Index: 446

Text #4

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.7081. Min Act: -0.0000

Data Index: 1146952 (C4 (Web Text))

Max Activating Token Index: 881

Text #5

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6801. Min Act: -0.0000

Data Index: 1015709 (C4 (Web Text))

Max Activating Token Index: 857

Text #6

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6539. Min Act: -0.0000

Data Index: 1323139 (C4 (Web Text))

Max Activating Token Index: 287

Text #7

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6927. Min Act: -0.0001

Data Index: 1125002 (C4 (Web Text))

Max Activating Token Index: 967

Text #8

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6851. Min Act: -0.0001

Data Index: 1263844 (C4 (Web Text))

Max Activating Token Index: 333

Text #9

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6582. Min Act: -0.0001

Data Index: 435735 (C4 (Web Text))

Max Activating Token Index: 983

Text #10

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6181. Min Act: -0.0000

Data Index: 1094989 (C4 (Web Text))

Max Activating Token Index: 255

Text #11

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6676. Min Act: -0.0001

Data Index: 389358 (C4 (Web Text))

Max Activating Token Index: 557

Text #12

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6250. Min Act: -0.0001

Data Index: 1057701 (C4 (Web Text))

Max Activating Token Index: 1012

Text #13

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6561. Min Act: -0.0001

Data Index: 1062725 (C4 (Web Text))

Max Activating Token Index: 997

Text #14

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6318. Min Act: -0.0000

Data Index: 99222 (C4 (Web Text))

Max Activating Token Index: 858

Text #15

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6477. Min Act: -0.0000

Data Index: 1589118 (Python Code)

Max Activating Token Index: 322

Text #16

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6386. Min Act: -0.0001

Data Index: 1006553 (C4 (Web Text))

Max Activating Token Index: 367

Text #17

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.5983. Min Act: -0.0000

Data Index: 132322 (C4 (Web Text))

Max Activating Token Index: 292

Text #18

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6191. Min Act: -0.0001

Data Index: 77080 (C4 (Web Text))

Max Activating Token Index: 622

Text #19

Max Range: 0.8226. Min Range: -0.8226

Max Act: 0.6017. Min Act: -0.0001

Data Index: 538914 (C4 (Web Text))

Max Activating Token Index: 273