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

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

Neuron 550 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: 1.9320. Min Range: -1.9320

Max Act: 1.9320. Min Act: -0.0000

Data Index: 1445952 (Python Code)

Max Activating Token Index: 202

Text #1

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.7719. Min Act: -0.0000

Data Index: 623647 (C4 (Web Text))

Max Activating Token Index: 197

Text #2

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.6858. Min Act: -0.0000

Data Index: 1022592 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.4974. Min Act: -0.0000

Data Index: 1251052 (C4 (Web Text))

Max Activating Token Index: 233

Text #4

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.6855. Min Act: -0.0000

Data Index: 1592041 (Python Code)

Max Activating Token Index: 351

Text #5

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.7520. Min Act: -0.0000

Data Index: 25156 (C4 (Web Text))

Max Activating Token Index: 383

Text #6

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.6069. Min Act: -0.0000

Data Index: 78826 (C4 (Web Text))

Max Activating Token Index: 914

Text #7

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.6022. Min Act: -0.0000

Data Index: 819698 (C4 (Web Text))

Max Activating Token Index: 364

Text #8

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.6763. Min Act: -0.0000

Data Index: 222009 (C4 (Web Text))

Max Activating Token Index: 761

Text #9

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5482. Min Act: -0.0000

Data Index: 268004 (C4 (Web Text))

Max Activating Token Index: 37

Text #10

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5411. Min Act: -0.0000

Data Index: 317346 (C4 (Web Text))

Max Activating Token Index: 49

Text #11

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5228. Min Act: -0.0000

Data Index: 584511 (C4 (Web Text))

Max Activating Token Index: 39

Text #12

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5908. Min Act: -0.0000

Data Index: 1411153 (Python Code)

Max Activating Token Index: 918

Text #13

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5960. Min Act: -0.0000

Data Index: 1034460 (C4 (Web Text))

Max Activating Token Index: 776

Text #14

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5175. Min Act: -0.0000

Data Index: 1329596 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5344. Min Act: -0.0000

Data Index: 372142 (C4 (Web Text))

Max Activating Token Index: 819

Text #16

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5247. Min Act: -0.0000

Data Index: 1007814 (C4 (Web Text))

Max Activating Token Index: 512

Text #17

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5847. Min Act: -0.0000

Data Index: 176981 (C4 (Web Text))

Max Activating Token Index: 900

Text #18

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.4113. Min Act: -0.0000

Data Index: 945175 (C4 (Web Text))

Max Activating Token Index: 137

Text #19

Max Range: 1.9320. Min Range: -1.9320

Max Act: 1.5746. Min Act: -0.0000

Data Index: 256337 (C4 (Web Text))

Max Activating Token Index: 247