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

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

Neuron 1815 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.4638. Min Act: -0.0001

Data Index: 445156 (C4 (Web Text))

Max Activating Token Index: 666

Text #1

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.4197. Min Act: -0.0001

Data Index: 122682 (C4 (Web Text))

Max Activating Token Index: 485

Text #2

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.3447. Min Act: -0.0001

Data Index: 549077 (C4 (Web Text))

Max Activating Token Index: 479

Text #3

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2982. Min Act: -0.0001

Data Index: 1246720 (C4 (Web Text))

Max Activating Token Index: 533

Text #4

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2692. Min Act: -0.0000

Data Index: 774364 (C4 (Web Text))

Max Activating Token Index: 133

Text #5

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2932. Min Act: -0.0000

Data Index: 629493 (C4 (Web Text))

Max Activating Token Index: 437

Text #6

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.3233. Min Act: -0.0001

Data Index: 758906 (C4 (Web Text))

Max Activating Token Index: 762

Text #7

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2823. Min Act: -0.0000

Data Index: 1613592 (Python Code)

Max Activating Token Index: 334

Text #8

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2274. Min Act: -0.0000

Data Index: 231677 (C4 (Web Text))

Max Activating Token Index: 887

Text #9

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2459. Min Act: -0.0000

Data Index: 1388765 (Python Code)

Max Activating Token Index: 342

Text #10

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2293. Min Act: -0.0000

Data Index: 1459752 (Python Code)

Max Activating Token Index: 441

Text #11

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2185. Min Act: -0.0001

Data Index: 1536995 (Python Code)

Max Activating Token Index: 760

Text #12

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2281. Min Act: -0.0000

Data Index: 1387833 (Python Code)

Max Activating Token Index: 542

Text #13

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2348. Min Act: -0.0001

Data Index: 458390 (C4 (Web Text))

Max Activating Token Index: 556

Text #14

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2203. Min Act: -0.0000

Data Index: 1503844 (Python Code)

Max Activating Token Index: 330

Text #15

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2024. Min Act: -0.0001

Data Index: 1431758 (Python Code)

Max Activating Token Index: 170

Text #16

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2246. Min Act: -0.0000

Data Index: 1428376 (Python Code)

Max Activating Token Index: 524

Text #17

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2002. Min Act: -0.0001

Data Index: 1594500 (Python Code)

Max Activating Token Index: 813

Text #18

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.2112. Min Act: -0.0000

Data Index: 1542661 (Python Code)

Max Activating Token Index: 508

Text #19

Max Range: 0.4638. Min Range: -0.4638

Max Act: 0.1880. Min Act: -0.0000

Data Index: 1412955 (Python Code)

Max Activating Token Index: 252