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

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

Neuron 1737 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: 0.4724. Min Range: -0.4724

Max Act: 0.4724. Min Act: -0.0001

Data Index: 418086 (C4 (Web Text))

Max Activating Token Index: 451

Text #1

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.4255. Min Act: -0.0001

Data Index: 1399927 (Python Code)

Max Activating Token Index: 413

Text #2

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.4197. Min Act: -0.0001

Data Index: 378408 (C4 (Web Text))

Max Activating Token Index: 41

Text #3

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.4199. Min Act: -0.0001

Data Index: 852206 (C4 (Web Text))

Max Activating Token Index: 88

Text #4

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3932. Min Act: -0.0001

Data Index: 261706 (C4 (Web Text))

Max Activating Token Index: 504

Text #5

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3954. Min Act: -0.0001

Data Index: 385315 (C4 (Web Text))

Max Activating Token Index: 636

Text #6

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3552. Min Act: -0.0001

Data Index: 57171 (C4 (Web Text))

Max Activating Token Index: 369

Text #7

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3605. Min Act: -0.0001

Data Index: 599363 (C4 (Web Text))

Max Activating Token Index: 784

Text #8

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.4007. Min Act: -0.0000

Data Index: 1108778 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3895. Min Act: -0.0001

Data Index: 677893 (C4 (Web Text))

Max Activating Token Index: 454

Text #10

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3992. Min Act: -0.0001

Data Index: 1159055 (C4 (Web Text))

Max Activating Token Index: 84

Text #11

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.4076. Min Act: -0.0000

Data Index: 862049 (C4 (Web Text))

Max Activating Token Index: 927

Text #12

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3948. Min Act: -0.0001

Data Index: 270436 (C4 (Web Text))

Max Activating Token Index: 679

Text #13

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3757. Min Act: -0.0001

Data Index: 736906 (C4 (Web Text))

Max Activating Token Index: 603

Text #14

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3852. Min Act: -0.0000

Data Index: 1151738 (C4 (Web Text))

Max Activating Token Index: 126

Text #15

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3642. Min Act: -0.0000

Data Index: 1116372 (C4 (Web Text))

Max Activating Token Index: 985

Text #16

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3793. Min Act: -0.0001

Data Index: 142295 (C4 (Web Text))

Max Activating Token Index: 358

Text #17

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3798. Min Act: -0.0001

Data Index: 1619685 (Python Code)

Max Activating Token Index: 996

Text #18

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3620. Min Act: -0.0000

Data Index: 1098239 (C4 (Web Text))

Max Activating Token Index: 742

Text #19

Max Range: 0.4724. Min Range: -0.4724

Max Act: 0.3847. Min Act: -0.0000

Data Index: 826687 (C4 (Web Text))

Max Activating Token Index: 969