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

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

Neuron 3107 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.5948. Min Act: -0.0000

Data Index: 1132813 (C4 (Web Text))

Max Activating Token Index: 589

Text #1

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.4096. Min Act: -0.0000

Data Index: 44689 (C4 (Web Text))

Max Activating Token Index: 910

Text #2

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.3791. Min Act: -0.0000

Data Index: 703454 (C4 (Web Text))

Max Activating Token Index: 325

Text #3

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.3231. Min Act: -0.0000

Data Index: 538674 (C4 (Web Text))

Max Activating Token Index: 837

Text #4

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.3243. Min Act: -0.0000

Data Index: 1191230 (C4 (Web Text))

Max Activating Token Index: 180

Text #5

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.3182. Min Act: -0.0000

Data Index: 435533 (C4 (Web Text))

Max Activating Token Index: 976

Text #6

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.3194. Min Act: -0.0000

Data Index: 568606 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2882. Min Act: -0.0000

Data Index: 588554 (C4 (Web Text))

Max Activating Token Index: 43

Text #8

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2844. Min Act: -0.0000

Data Index: 514742 (C4 (Web Text))

Max Activating Token Index: 504

Text #9

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2731. Min Act: -0.0000

Data Index: 1281157 (C4 (Web Text))

Max Activating Token Index: 1001

Text #10

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2794. Min Act: -0.0000

Data Index: 1514091 (Python Code)

Max Activating Token Index: 457

Text #11

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2685. Min Act: -0.0000

Data Index: 378185 (C4 (Web Text))

Max Activating Token Index: 52

Text #12

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2741. Min Act: -0.0000

Data Index: 309192 (C4 (Web Text))

Max Activating Token Index: 1004

Text #13

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2583. Min Act: -0.0000

Data Index: 1617319 (Python Code)

Max Activating Token Index: 279

Text #14

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2739. Min Act: -0.0000

Data Index: 1293238 (C4 (Web Text))

Max Activating Token Index: 790

Text #15

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2618. Min Act: -0.0000

Data Index: 253355 (C4 (Web Text))

Max Activating Token Index: 166

Text #16

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2576. Min Act: -0.0000

Data Index: 379363 (C4 (Web Text))

Max Activating Token Index: 545

Text #17

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2411. Min Act: -0.0000

Data Index: 774453 (C4 (Web Text))

Max Activating Token Index: 915

Text #18

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2324. Min Act: -0.0000

Data Index: 702115 (C4 (Web Text))

Max Activating Token Index: 421

Text #19

Max Range: 0.5948. Min Range: -0.5948

Max Act: 0.2556. Min Act: -0.0000

Data Index: 19835 (C4 (Web Text))

Max Activating Token Index: 961