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

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

Neuron 4090 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0404. Min Act: -0.0001

Data Index: 386566 (C4 (Web Text))

Max Activating Token Index: 463

Text #1

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0378. Min Act: -0.0001

Data Index: 867743 (C4 (Web Text))

Max Activating Token Index: 23

Text #2

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0372. Min Act: -0.0001

Data Index: 1479055 (Python Code)

Max Activating Token Index: 480

Text #3

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0353. Min Act: -0.0001

Data Index: 877304 (C4 (Web Text))

Max Activating Token Index: 448

Text #4

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0352. Min Act: -0.0001

Data Index: 226197 (C4 (Web Text))

Max Activating Token Index: 65

Text #5

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0341. Min Act: -0.0001

Data Index: 1300990 (C4 (Web Text))

Max Activating Token Index: 470

Text #6

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0341. Min Act: -0.0001

Data Index: 1050165 (C4 (Web Text))

Max Activating Token Index: 623

Text #7

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0345. Min Act: -0.0001

Data Index: 433688 (C4 (Web Text))

Max Activating Token Index: 641

Text #8

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0338. Min Act: -0.0001

Data Index: 949392 (C4 (Web Text))

Max Activating Token Index: 317

Text #9

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0316. Min Act: -0.0001

Data Index: 1470637 (Python Code)

Max Activating Token Index: 299

Text #10

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0335. Min Act: -0.0001

Data Index: 243665 (C4 (Web Text))

Max Activating Token Index: 234

Text #11

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0318. Min Act: -0.0001

Data Index: 1348973 (C4 (Web Text))

Max Activating Token Index: 819

Text #12

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0307. Min Act: -0.0001

Data Index: 413290 (C4 (Web Text))

Max Activating Token Index: 599

Text #13

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0321. Min Act: -0.0001

Data Index: 444221 (C4 (Web Text))

Max Activating Token Index: 37

Text #14

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0316. Min Act: -0.0001

Data Index: 131960 (C4 (Web Text))

Max Activating Token Index: 45

Text #15

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0301. Min Act: -0.0001

Data Index: 1455938 (Python Code)

Max Activating Token Index: 830

Text #16

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0292. Min Act: -0.0001

Data Index: 287377 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0281. Min Act: -0.0001

Data Index: 752423 (C4 (Web Text))

Max Activating Token Index: 292

Text #18

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0292. Min Act: -0.0001

Data Index: 1172956 (C4 (Web Text))

Max Activating Token Index: 26

Text #19

Max Range: 0.0404. Min Range: -0.0404

Max Act: 0.0281. Min Act: -0.0001

Data Index: 980396 (C4 (Web Text))

Max Activating Token Index: 826