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

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

Neuron 1047 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: 2.3443. Min Range: -2.3443

Max Act: 2.3443. Min Act: -0.0000

Data Index: 211601 (C4 (Web Text))

Max Activating Token Index: 549

Text #1

Max Range: 2.3443. Min Range: -2.3443

Max Act: 2.0234. Min Act: -0.0000

Data Index: 937931 (C4 (Web Text))

Max Activating Token Index: 558

Text #2

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.8516. Min Act: -0.0000

Data Index: 186805 (C4 (Web Text))

Max Activating Token Index: 535

Text #3

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.8173. Min Act: -0.0000

Data Index: 490624 (C4 (Web Text))

Max Activating Token Index: 49

Text #4

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.5822. Min Act: -0.0000

Data Index: 1440407 (Python Code)

Max Activating Token Index: 685

Text #5

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.5978. Min Act: -0.0000

Data Index: 1474217 (Python Code)

Max Activating Token Index: 197

Text #6

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.4367. Min Act: -0.0000

Data Index: 246624 (C4 (Web Text))

Max Activating Token Index: 934

Text #7

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.5041. Min Act: -0.0000

Data Index: 1445659 (Python Code)

Max Activating Token Index: 878

Text #8

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.5912. Min Act: -0.0000

Data Index: 53638 (C4 (Web Text))

Max Activating Token Index: 407

Text #9

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.4877. Min Act: -0.0000

Data Index: 549824 (C4 (Web Text))

Max Activating Token Index: 533

Text #10

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.6345. Min Act: -0.0000

Data Index: 498443 (C4 (Web Text))

Max Activating Token Index: 781

Text #11

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3040. Min Act: -0.0000

Data Index: 355146 (C4 (Web Text))

Max Activating Token Index: 224

Text #12

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3539. Min Act: -0.0000

Data Index: 641074 (C4 (Web Text))

Max Activating Token Index: 168

Text #13

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3507. Min Act: -0.0000

Data Index: 254401 (C4 (Web Text))

Max Activating Token Index: 588

Text #14

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3320. Min Act: -0.0000

Data Index: 1436873 (Python Code)

Max Activating Token Index: 536

Text #15

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3890. Min Act: -0.0000

Data Index: 127685 (C4 (Web Text))

Max Activating Token Index: 302

Text #16

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3025. Min Act: -0.0000

Data Index: 1259970 (C4 (Web Text))

Max Activating Token Index: 426

Text #17

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.2837. Min Act: -0.0000

Data Index: 1274650 (C4 (Web Text))

Max Activating Token Index: 917

Text #18

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.2904. Min Act: -0.0000

Data Index: 1229015 (C4 (Web Text))

Max Activating Token Index: 279

Text #19

Max Range: 2.3443. Min Range: -2.3443

Max Act: 1.3646. Min Act: -0.0000

Data Index: 360369 (C4 (Web Text))

Max Activating Token Index: 214