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

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

Neuron 627 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.3088. Min Act: -0.0002

Data Index: 339694 (C4 (Web Text))

Max Activating Token Index: 257

Text #1

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2775. Min Act: -0.0002

Data Index: 793543 (C4 (Web Text))

Max Activating Token Index: 943

Text #2

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2890. Min Act: -0.0002

Data Index: 327308 (C4 (Web Text))

Max Activating Token Index: 908

Text #3

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2714. Min Act: -0.0002

Data Index: 530780 (C4 (Web Text))

Max Activating Token Index: 505

Text #4

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2615. Min Act: -0.0002

Data Index: 1442072 (Python Code)

Max Activating Token Index: 865

Text #5

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2627. Min Act: -0.0002

Data Index: 559688 (C4 (Web Text))

Max Activating Token Index: 864

Text #6

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2637. Min Act: -0.0002

Data Index: 552896 (C4 (Web Text))

Max Activating Token Index: 672

Text #7

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2428. Min Act: -0.0002

Data Index: 1399784 (Python Code)

Max Activating Token Index: 140

Text #8

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2441. Min Act: -0.0002

Data Index: 900338 (C4 (Web Text))

Max Activating Token Index: 645

Text #9

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2422. Min Act: -0.0002

Data Index: 1232804 (C4 (Web Text))

Max Activating Token Index: 832

Text #10

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2499. Min Act: -0.0002

Data Index: 86065 (C4 (Web Text))

Max Activating Token Index: 769

Text #11

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2365. Min Act: -0.0002

Data Index: 106234 (C4 (Web Text))

Max Activating Token Index: 392

Text #12

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2292. Min Act: -0.0002

Data Index: 377592 (C4 (Web Text))

Max Activating Token Index: 260

Text #13

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2428. Min Act: -0.0002

Data Index: 46211 (C4 (Web Text))

Max Activating Token Index: 727

Text #14

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2410. Min Act: -0.0002

Data Index: 1565544 (Python Code)

Max Activating Token Index: 650

Text #15

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2340. Min Act: -0.0002

Data Index: 983326 (C4 (Web Text))

Max Activating Token Index: 236

Text #16

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2248. Min Act: -0.0002

Data Index: 761479 (C4 (Web Text))

Max Activating Token Index: 721

Text #17

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2256. Min Act: -0.0002

Data Index: 1183987 (C4 (Web Text))

Max Activating Token Index: 74

Text #18

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2194. Min Act: -0.0002

Data Index: 728015 (C4 (Web Text))

Max Activating Token Index: 937

Text #19

Max Range: 0.3088. Min Range: -0.3088

Max Act: 0.2193. Min Act: -0.0002

Data Index: 1069771 (C4 (Web Text))

Max Activating Token Index: 383