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

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

Neuron 5006 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.9756. Min Act: -0.0000

Data Index: 1552814 (Python Code)

Max Activating Token Index: 898

Text #1

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.7308. Min Act: -0.0000

Data Index: 1066095 (C4 (Web Text))

Max Activating Token Index: 701

Text #2

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.3368. Min Act: -0.0000

Data Index: 1501459 (Python Code)

Max Activating Token Index: 621

Text #3

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.3662. Min Act: -0.0000

Data Index: 490900 (C4 (Web Text))

Max Activating Token Index: 634

Text #4

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.3647. Min Act: -0.0000

Data Index: 1648933 (Python Code)

Max Activating Token Index: 661

Text #5

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.2791. Min Act: -0.0000

Data Index: 1399430 (Python Code)

Max Activating Token Index: 343

Text #6

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.2391. Min Act: -0.0000

Data Index: 1119127 (C4 (Web Text))

Max Activating Token Index: 647

Text #7

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.0555. Min Act: -0.0000

Data Index: 1391814 (Python Code)

Max Activating Token Index: 13

Text #8

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.8844. Min Act: -0.0000

Data Index: 1047310 (C4 (Web Text))

Max Activating Token Index: 903

Text #9

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.8493. Min Act: -0.0000

Data Index: 1359864 (Python Code)

Max Activating Token Index: 192

Text #10

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.9385. Min Act: -0.0000

Data Index: 967698 (C4 (Web Text))

Max Activating Token Index: 382

Text #11

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.7597. Min Act: -0.0000

Data Index: 162111 (C4 (Web Text))

Max Activating Token Index: 440

Text #12

Max Range: 3.9756. Min Range: -3.9756

Max Act: 3.0322. Min Act: -0.0000

Data Index: 726430 (C4 (Web Text))

Max Activating Token Index: 493

Text #13

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.7152. Min Act: -0.0000

Data Index: 775731 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.9885. Min Act: -0.0000

Data Index: 577689 (C4 (Web Text))

Max Activating Token Index: 430

Text #15

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.7230. Min Act: -0.0000

Data Index: 443806 (C4 (Web Text))

Max Activating Token Index: 191

Text #16

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.8680. Min Act: -0.0000

Data Index: 1633625 (Python Code)

Max Activating Token Index: 30

Text #17

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.7082. Min Act: -0.0000

Data Index: 628097 (C4 (Web Text))

Max Activating Token Index: 232

Text #18

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.6917. Min Act: -0.0000

Data Index: 1234440 (C4 (Web Text))

Max Activating Token Index: 530

Text #19

Max Range: 3.9756. Min Range: -3.9756

Max Act: 2.6505. Min Act: -0.0000

Data Index: 1543020 (Python Code)

Max Activating Token Index: 340