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

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

Neuron 3270 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1394. Min Range: -1.1394

Max Act: 1.1394. Min Act: -0.0000

Data Index: 1411841 (Python Code)

Max Activating Token Index: 25

Text #1

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.9731. Min Act: -0.0000

Data Index: 742940 (C4 (Web Text))

Max Activating Token Index: 90

Text #2

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.8516. Min Act: -0.0000

Data Index: 1622015 (Python Code)

Max Activating Token Index: 230

Text #3

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.9089. Min Act: -0.0000

Data Index: 364964 (C4 (Web Text))

Max Activating Token Index: 196

Text #4

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.8468. Min Act: -0.0000

Data Index: 767887 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.8340. Min Act: -0.0000

Data Index: 1357606 (C4 (Web Text))

Max Activating Token Index: 380

Text #6

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.8223. Min Act: -0.0000

Data Index: 1045198 (C4 (Web Text))

Max Activating Token Index: 266

Text #7

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7916. Min Act: -0.0000

Data Index: 438286 (C4 (Web Text))

Max Activating Token Index: 358

Text #8

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7319. Min Act: -0.0000

Data Index: 1235127 (C4 (Web Text))

Max Activating Token Index: 567

Text #9

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7419. Min Act: -0.0000

Data Index: 1496013 (Python Code)

Max Activating Token Index: 318

Text #10

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7979. Min Act: -0.0000

Data Index: 1636729 (Python Code)

Max Activating Token Index: 164

Text #11

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7559. Min Act: -0.0000

Data Index: 1272125 (C4 (Web Text))

Max Activating Token Index: 356

Text #12

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7767. Min Act: -0.0000

Data Index: 1653779 (Python Code)

Max Activating Token Index: 23

Text #13

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7253. Min Act: -0.0000

Data Index: 1280779 (C4 (Web Text))

Max Activating Token Index: 346

Text #14

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7870. Min Act: -0.0000

Data Index: 1611508 (Python Code)

Max Activating Token Index: 626

Text #15

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7488. Min Act: -0.0000

Data Index: 1647587 (Python Code)

Max Activating Token Index: 533

Text #16

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7213. Min Act: -0.0000

Data Index: 1574326 (Python Code)

Max Activating Token Index: 31

Text #17

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7454. Min Act: -0.0000

Data Index: 1449377 (Python Code)

Max Activating Token Index: 769

Text #18

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.6946. Min Act: -0.0000

Data Index: 1602929 (Python Code)

Max Activating Token Index: 355

Text #19

Max Range: 1.1394. Min Range: -1.1394

Max Act: 0.7356. Min Act: -0.0000

Data Index: 1455689 (Python Code)

Max Activating Token Index: 547