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

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

Neuron 5716 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.4088. Min Act: -0.0000

Data Index: 569183 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.3106. Min Act: -0.0000

Data Index: 704810 (C4 (Web Text))

Max Activating Token Index: 825

Text #2

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2705. Min Act: -0.0000

Data Index: 594756 (C4 (Web Text))

Max Activating Token Index: 208

Text #3

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2661. Min Act: -0.0000

Data Index: 1232305 (C4 (Web Text))

Max Activating Token Index: 182

Text #4

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2744. Min Act: -0.0000

Data Index: 1629115 (Python Code)

Max Activating Token Index: 271

Text #5

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2719. Min Act: -0.0000

Data Index: 605760 (C4 (Web Text))

Max Activating Token Index: 742

Text #6

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2784. Min Act: -0.0000

Data Index: 718705 (C4 (Web Text))

Max Activating Token Index: 111

Text #7

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2587. Min Act: -0.0000

Data Index: 122502 (C4 (Web Text))

Max Activating Token Index: 792

Text #8

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2755. Min Act: -0.0000

Data Index: 213373 (C4 (Web Text))

Max Activating Token Index: 107

Text #9

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2444. Min Act: -0.0000

Data Index: 719624 (C4 (Web Text))

Max Activating Token Index: 233

Text #10

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2846. Min Act: -0.0000

Data Index: 747608 (C4 (Web Text))

Max Activating Token Index: 970

Text #11

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2537. Min Act: -0.0000

Data Index: 725611 (C4 (Web Text))

Max Activating Token Index: 208

Text #12

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2527. Min Act: -0.0000

Data Index: 1574531 (Python Code)

Max Activating Token Index: 497

Text #13

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2369. Min Act: -0.0000

Data Index: 196377 (C4 (Web Text))

Max Activating Token Index: 239

Text #14

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2919. Min Act: -0.0000

Data Index: 65000 (C4 (Web Text))

Max Activating Token Index: 348

Text #15

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2402. Min Act: -0.0000

Data Index: 242103 (C4 (Web Text))

Max Activating Token Index: 354

Text #16

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2255. Min Act: -0.0000

Data Index: 75386 (C4 (Web Text))

Max Activating Token Index: 196

Text #17

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2334. Min Act: -0.0000

Data Index: 1030891 (C4 (Web Text))

Max Activating Token Index: 563

Text #18

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2420. Min Act: -0.0000

Data Index: 497256 (C4 (Web Text))

Max Activating Token Index: 601

Text #19

Max Range: 0.4088. Min Range: -0.4088

Max Act: 0.2497. Min Act: -0.0000

Data Index: 1032570 (C4 (Web Text))

Max Activating Token Index: 766