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

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

Neuron 312 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: 0.0555. Min Range: -0.0555

Max Act: 0.0555. Min Act: -0.0002

Data Index: 1218114 (C4 (Web Text))

Max Activating Token Index: 542

Text #1

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0547. Min Act: -0.0002

Data Index: 192970 (C4 (Web Text))

Max Activating Token Index: 654

Text #2

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0545. Min Act: -0.0001

Data Index: 435114 (C4 (Web Text))

Max Activating Token Index: 8

Text #3

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0499. Min Act: -0.0002

Data Index: 483994 (C4 (Web Text))

Max Activating Token Index: 153

Text #4

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0478. Min Act: -0.0002

Data Index: 393979 (C4 (Web Text))

Max Activating Token Index: 170

Text #5

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0448. Min Act: -0.0002

Data Index: 353935 (C4 (Web Text))

Max Activating Token Index: 244

Text #6

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0455. Min Act: -0.0001

Data Index: 445517 (C4 (Web Text))

Max Activating Token Index: 44

Text #7

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0449. Min Act: -0.0002

Data Index: 671130 (C4 (Web Text))

Max Activating Token Index: 44

Text #8

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0422. Min Act: -0.0002

Data Index: 1265605 (C4 (Web Text))

Max Activating Token Index: 518

Text #9

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0427. Min Act: -0.0001

Data Index: 1051632 (C4 (Web Text))

Max Activating Token Index: 984

Text #10

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0439. Min Act: -0.0002

Data Index: 395163 (C4 (Web Text))

Max Activating Token Index: 305

Text #11

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0406. Min Act: -0.0002

Data Index: 703397 (C4 (Web Text))

Max Activating Token Index: 58

Text #12

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0404. Min Act: -0.0002

Data Index: 765279 (C4 (Web Text))

Max Activating Token Index: 255

Text #13

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0427. Min Act: -0.0001

Data Index: 692679 (C4 (Web Text))

Max Activating Token Index: 779

Text #14

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0416. Min Act: -0.0002

Data Index: 91330 (C4 (Web Text))

Max Activating Token Index: 717

Text #15

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0394. Min Act: -0.0002

Data Index: 1152993 (C4 (Web Text))

Max Activating Token Index: 422

Text #16

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0395. Min Act: -0.0002

Data Index: 855989 (C4 (Web Text))

Max Activating Token Index: 333

Text #17

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0389. Min Act: -0.0002

Data Index: 199552 (C4 (Web Text))

Max Activating Token Index: 184

Text #18

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0415. Min Act: -0.0002

Data Index: 1023053 (C4 (Web Text))

Max Activating Token Index: 388

Text #19

Max Range: 0.0555. Min Range: -0.0555

Max Act: 0.0382. Min Act: -0.0002

Data Index: 612160 (C4 (Web Text))

Max Activating Token Index: 864