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

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

Neuron 3637 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: 0.3404. Min Range: -0.3404

Max Act: 0.3404. Min Act: -0.0000

Data Index: 1454688 (Python Code)

Max Activating Token Index: 574

Text #1

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.3150. Min Act: -0.0000

Data Index: 1070233 (C4 (Web Text))

Max Activating Token Index: 93

Text #2

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.3258. Min Act: -0.0000

Data Index: 1444398 (Python Code)

Max Activating Token Index: 720

Text #3

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.3138. Min Act: -0.0000

Data Index: 1403183 (Python Code)

Max Activating Token Index: 242

Text #4

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.3070. Min Act: -0.0000

Data Index: 1534326 (Python Code)

Max Activating Token Index: 353

Text #5

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2871. Min Act: -0.0000

Data Index: 1404008 (Python Code)

Max Activating Token Index: 601

Text #6

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2936. Min Act: -0.0000

Data Index: 751802 (C4 (Web Text))

Max Activating Token Index: 17

Text #7

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2758. Min Act: -0.0000

Data Index: 1454862 (Python Code)

Max Activating Token Index: 416

Text #8

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2651. Min Act: -0.0000

Data Index: 1413784 (Python Code)

Max Activating Token Index: 199

Text #9

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2626. Min Act: -0.0000

Data Index: 1280857 (C4 (Web Text))

Max Activating Token Index: 713

Text #10

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2838. Min Act: -0.0000

Data Index: 1650501 (Python Code)

Max Activating Token Index: 363

Text #11

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2636. Min Act: -0.0000

Data Index: 1526359 (Python Code)

Max Activating Token Index: 462

Text #12

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2608. Min Act: -0.0000

Data Index: 1370362 (Python Code)

Max Activating Token Index: 714

Text #13

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2488. Min Act: -0.0000

Data Index: 1442692 (Python Code)

Max Activating Token Index: 822

Text #14

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2372. Min Act: -0.0000

Data Index: 1504052 (Python Code)

Max Activating Token Index: 947

Text #15

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2463. Min Act: -0.0000

Data Index: 1394694 (Python Code)

Max Activating Token Index: 741

Text #16

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2314. Min Act: -0.0000

Data Index: 438574 (C4 (Web Text))

Max Activating Token Index: 298

Text #17

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2388. Min Act: -0.0000

Data Index: 1628045 (Python Code)

Max Activating Token Index: 671

Text #18

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2473. Min Act: -0.0000

Data Index: 689161 (C4 (Web Text))

Max Activating Token Index: 763

Text #19

Max Range: 0.3404. Min Range: -0.3404

Max Act: 0.2464. Min Act: -0.0000

Data Index: 385900 (C4 (Web Text))

Max Activating Token Index: 657