Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1921 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0414. Min Act: -0.0002

Data Index: 953785 (C4 (Web Text))

Max Activating Token Index: 826

Text #1

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0389. Min Act: -0.0001

Data Index: 655686 (C4 (Web Text))

Max Activating Token Index: 732

Text #2

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0382. Min Act: -0.0002

Data Index: 761144 (C4 (Web Text))

Max Activating Token Index: 742

Text #3

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0378. Min Act: -0.0002

Data Index: 834457 (C4 (Web Text))

Max Activating Token Index: 415

Text #4

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0376. Min Act: -0.0002

Data Index: 790237 (C4 (Web Text))

Max Activating Token Index: 353

Text #5

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0369. Min Act: -0.0002

Data Index: 1402817 (Python Code)

Max Activating Token Index: 906

Text #6

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0376. Min Act: -0.0002

Data Index: 1633890 (Python Code)

Max Activating Token Index: 583

Text #7

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0367. Min Act: -0.0001

Data Index: 1357267 (C4 (Web Text))

Max Activating Token Index: 514

Text #8

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0361. Min Act: -0.0002

Data Index: 1637961 (Python Code)

Max Activating Token Index: 325

Text #9

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0348. Min Act: -0.0002

Data Index: 492041 (C4 (Web Text))

Max Activating Token Index: 455

Text #10

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0349. Min Act: -0.0001

Data Index: 43578 (C4 (Web Text))

Max Activating Token Index: 496

Text #11

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0363. Min Act: -0.0002

Data Index: 1183654 (C4 (Web Text))

Max Activating Token Index: 668

Text #12

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0347. Min Act: -0.0001

Data Index: 1397771 (Python Code)

Max Activating Token Index: 898

Text #13

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0359. Min Act: -0.0002

Data Index: 1276024 (C4 (Web Text))

Max Activating Token Index: 16

Text #14

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0352. Min Act: -0.0002

Data Index: 1093405 (C4 (Web Text))

Max Activating Token Index: 339

Text #15

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0349. Min Act: -0.0001

Data Index: 1364699 (Python Code)

Max Activating Token Index: 311

Text #16

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0344. Min Act: -0.0002

Data Index: 1049007 (C4 (Web Text))

Max Activating Token Index: 426

Text #17

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0332. Min Act: -0.0002

Data Index: 424727 (C4 (Web Text))

Max Activating Token Index: 805

Text #18

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0337. Min Act: -0.0002

Data Index: 420196 (C4 (Web Text))

Max Activating Token Index: 305

Text #19

Max Range: 0.0414. Min Range: -0.0414

Max Act: 0.0313. Min Act: -0.0001

Data Index: 1542048 (Python Code)

Max Activating Token Index: 1017