Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 938 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.4464. Min Act: 0.0000

Data Index: 498281 (C4 (Web Text))

Max Activating Token Index: 622

Text #1

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.2373. Min Act: -0.0000

Data Index: 37996 (C4 (Web Text))

Max Activating Token Index: 453

Text #2

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.1806. Min Act: -0.0000

Data Index: 72388 (C4 (Web Text))

Max Activating Token Index: 642

Text #3

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.1606. Min Act: 0.0000

Data Index: 152604 (C4 (Web Text))

Max Activating Token Index: 263

Text #4

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.0961. Min Act: -0.0000

Data Index: 827952 (C4 (Web Text))

Max Activating Token Index: 540

Text #5

Max Range: 1.4464. Min Range: -1.4464

Max Act: 1.0539. Min Act: -0.0000

Data Index: 1111352 (C4 (Web Text))

Max Activating Token Index: 423

Text #6

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.9785. Min Act: -0.0000

Data Index: 1322440 (C4 (Web Text))

Max Activating Token Index: 1022

Text #7

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.8654. Min Act: -0.0001

Data Index: 501347 (C4 (Web Text))

Max Activating Token Index: 454

Text #8

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7996. Min Act: -0.0000

Data Index: 261374 (C4 (Web Text))

Max Activating Token Index: 878

Text #9

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.8169. Min Act: 0.0000

Data Index: 86611 (C4 (Web Text))

Max Activating Token Index: 260

Text #10

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7581. Min Act: -0.0000

Data Index: 942857 (C4 (Web Text))

Max Activating Token Index: 765

Text #11

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7769. Min Act: -0.0000

Data Index: 655412 (C4 (Web Text))

Max Activating Token Index: 811

Text #12

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7909. Min Act: -0.0001

Data Index: 1469504 (Python Code)

Max Activating Token Index: 93

Text #13

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7771. Min Act: -0.0001

Data Index: 1209529 (C4 (Web Text))

Max Activating Token Index: 86

Text #14

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7694. Min Act: -0.0000

Data Index: 125500 (C4 (Web Text))

Max Activating Token Index: 548

Text #15

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7602. Min Act: -0.0000

Data Index: 1134990 (C4 (Web Text))

Max Activating Token Index: 460

Text #16

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7422. Min Act: -0.0000

Data Index: 955976 (C4 (Web Text))

Max Activating Token Index: 357

Text #17

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7293. Min Act: -0.0001

Data Index: 600154 (C4 (Web Text))

Max Activating Token Index: 818

Text #18

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.7516. Min Act: -0.0000

Data Index: 180363 (C4 (Web Text))

Max Activating Token Index: 124

Text #19

Max Range: 1.4464. Min Range: -1.4464

Max Act: 0.6811. Min Act: -0.0000

Data Index: 949573 (C4 (Web Text))

Max Activating Token Index: 323