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

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

Neuron 1844 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2248. Min Range: -1.2248

Max Act: 1.2248. Min Act: -0.0000

Data Index: 987244 (C4 (Web Text))

Max Activating Token Index: 765

Text #1

Max Range: 1.2248. Min Range: -1.2248

Max Act: 1.1168. Min Act: -0.0000

Data Index: 990593 (C4 (Web Text))

Max Activating Token Index: 482

Text #2

Max Range: 1.2248. Min Range: -1.2248

Max Act: 1.0072. Min Act: -0.0000

Data Index: 837391 (C4 (Web Text))

Max Activating Token Index: 417

Text #3

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.8785. Min Act: -0.0000

Data Index: 632524 (C4 (Web Text))

Max Activating Token Index: 153

Text #4

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.8759. Min Act: -0.0000

Data Index: 389618 (C4 (Web Text))

Max Activating Token Index: 122

Text #5

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.8156. Min Act: -0.0000

Data Index: 1230511 (C4 (Web Text))

Max Activating Token Index: 368

Text #6

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.7918. Min Act: -0.0000

Data Index: 1018591 (C4 (Web Text))

Max Activating Token Index: 375

Text #7

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.7278. Min Act: -0.0000

Data Index: 1304646 (C4 (Web Text))

Max Activating Token Index: 69

Text #8

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.7671. Min Act: -0.0000

Data Index: 657906 (C4 (Web Text))

Max Activating Token Index: 162

Text #9

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.7161. Min Act: -0.0000

Data Index: 1125957 (C4 (Web Text))

Max Activating Token Index: 950

Text #10

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6987. Min Act: -0.0000

Data Index: 1440947 (Python Code)

Max Activating Token Index: 760

Text #11

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6569. Min Act: -0.0000

Data Index: 877382 (C4 (Web Text))

Max Activating Token Index: 849

Text #12

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6241. Min Act: -0.0000

Data Index: 666187 (C4 (Web Text))

Max Activating Token Index: 208

Text #13

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.7260. Min Act: -0.0000

Data Index: 799672 (C4 (Web Text))

Max Activating Token Index: 323

Text #14

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6585. Min Act: -0.0000

Data Index: 737939 (C4 (Web Text))

Max Activating Token Index: 568

Text #15

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6762. Min Act: -0.0000

Data Index: 625213 (C4 (Web Text))

Max Activating Token Index: 85

Text #16

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6610. Min Act: -0.0000

Data Index: 1373489 (Python Code)

Max Activating Token Index: 23

Text #17

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6079. Min Act: -0.0000

Data Index: 1603764 (Python Code)

Max Activating Token Index: 352

Text #18

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.6525. Min Act: -0.0000

Data Index: 728750 (C4 (Web Text))

Max Activating Token Index: 925

Text #19

Max Range: 1.2248. Min Range: -1.2248

Max Act: 0.5655. Min Act: -0.0000

Data Index: 1220785 (C4 (Web Text))

Max Activating Token Index: 332