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

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

Neuron 1108 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.8233. Min Act: -0.0000

Data Index: 423551 (C4 (Web Text))

Max Activating Token Index: 101

Text #1

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.6408. Min Act: -0.0000

Data Index: 266021 (C4 (Web Text))

Max Activating Token Index: 622

Text #2

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.5667. Min Act: -0.0000

Data Index: 572421 (C4 (Web Text))

Max Activating Token Index: 601

Text #3

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.2087. Min Act: -0.0000

Data Index: 561937 (C4 (Web Text))

Max Activating Token Index: 492

Text #4

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.4139. Min Act: -0.0000

Data Index: 1014779 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.2287. Min Act: -0.0000

Data Index: 424031 (C4 (Web Text))

Max Activating Token Index: 29

Text #6

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.1727. Min Act: -0.0000

Data Index: 1244822 (C4 (Web Text))

Max Activating Token Index: 44

Text #7

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.1720. Min Act: -0.0000

Data Index: 1347182 (C4 (Web Text))

Max Activating Token Index: 349

Text #8

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.1534. Min Act: -0.0000

Data Index: 139100 (C4 (Web Text))

Max Activating Token Index: 84

Text #9

Max Range: 3.8233. Min Range: -3.8233

Max Act: 3.2667. Min Act: -0.0000

Data Index: 505347 (C4 (Web Text))

Max Activating Token Index: 384

Text #10

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.9481. Min Act: -0.0000

Data Index: 1070027 (C4 (Web Text))

Max Activating Token Index: 614

Text #11

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.8180. Min Act: -0.0000

Data Index: 328186 (C4 (Web Text))

Max Activating Token Index: 239

Text #12

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.8959. Min Act: -0.0000

Data Index: 933986 (C4 (Web Text))

Max Activating Token Index: 452

Text #13

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.9432. Min Act: -0.0000

Data Index: 1266208 (C4 (Web Text))

Max Activating Token Index: 221

Text #14

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.6584. Min Act: -0.0000

Data Index: 1098180 (C4 (Web Text))

Max Activating Token Index: 265

Text #15

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.7687. Min Act: -0.0000

Data Index: 393850 (C4 (Web Text))

Max Activating Token Index: 95

Text #16

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.7537. Min Act: -0.0000

Data Index: 1297462 (C4 (Web Text))

Max Activating Token Index: 486

Text #17

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.4240. Min Act: -0.0000

Data Index: 1082449 (C4 (Web Text))

Max Activating Token Index: 851

Text #18

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.4531. Min Act: -0.0000

Data Index: 418299 (C4 (Web Text))

Max Activating Token Index: 608

Text #19

Max Range: 3.8233. Min Range: -3.8233

Max Act: 2.6750. Min Act: -0.0000

Data Index: 895323 (C4 (Web Text))

Max Activating Token Index: 683