Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2022 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8902. Min Act: -0.0000

Data Index: 431398 (C4 (Web Text))

Max Activating Token Index: 1013

Text #1

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8902. Min Act: -0.0000

Data Index: 423207 (C4 (Web Text))

Max Activating Token Index: 573

Text #2

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8549. Min Act: -0.0000

Data Index: 499077 (C4 (Web Text))

Max Activating Token Index: 576

Text #3

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8240. Min Act: -0.0000

Data Index: 74638 (C4 (Web Text))

Max Activating Token Index: 75

Text #4

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8149. Min Act: -0.0000

Data Index: 1161024 (C4 (Web Text))

Max Activating Token Index: 640

Text #5

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7786. Min Act: -0.0000

Data Index: 1108050 (C4 (Web Text))

Max Activating Token Index: 848

Text #6

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8018. Min Act: -0.0000

Data Index: 101972 (C4 (Web Text))

Max Activating Token Index: 89

Text #7

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7709. Min Act: -0.0000

Data Index: 1116203 (C4 (Web Text))

Max Activating Token Index: 529

Text #8

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7862. Min Act: -0.0000

Data Index: 411285 (C4 (Web Text))

Max Activating Token Index: 594

Text #9

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8168. Min Act: -0.0000

Data Index: 236873 (C4 (Web Text))

Max Activating Token Index: 935

Text #10

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.8097. Min Act: -0.0000

Data Index: 1153412 (C4 (Web Text))

Max Activating Token Index: 361

Text #11

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7987. Min Act: -0.0000

Data Index: 197632 (C4 (Web Text))

Max Activating Token Index: 647

Text #12

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7680. Min Act: -0.0000

Data Index: 1128433 (C4 (Web Text))

Max Activating Token Index: 626

Text #13

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7540. Min Act: -0.0000

Data Index: 407342 (C4 (Web Text))

Max Activating Token Index: 212

Text #14

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7428. Min Act: -0.0001

Data Index: 87226 (C4 (Web Text))

Max Activating Token Index: 47

Text #15

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7204. Min Act: -0.0000

Data Index: 1336590 (C4 (Web Text))

Max Activating Token Index: 748

Text #16

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7424. Min Act: -0.0000

Data Index: 412080 (C4 (Web Text))

Max Activating Token Index: 698

Text #17

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7343. Min Act: -0.0000

Data Index: 1195788 (C4 (Web Text))

Max Activating Token Index: 763

Text #18

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7303. Min Act: -0.0000

Data Index: 664913 (C4 (Web Text))

Max Activating Token Index: 218

Text #19

Max Range: 0.8902. Min Range: -0.8902

Max Act: 0.7743. Min Act: -0.0000

Data Index: 145320 (C4 (Web Text))

Max Activating Token Index: 565