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

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

Neuron 2020 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5804. Min Act: -0.0002

Data Index: 1284717 (C4 (Web Text))

Max Activating Token Index: 860

Text #1

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5665. Min Act: -0.0001

Data Index: 437678 (C4 (Web Text))

Max Activating Token Index: 985

Text #2

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5859. Min Act: -0.0002

Data Index: 338837 (C4 (Web Text))

Max Activating Token Index: 877

Text #3

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5450. Min Act: -0.0002

Data Index: 848271 (C4 (Web Text))

Max Activating Token Index: 950

Text #4

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5653. Min Act: -0.0002

Data Index: 562721 (C4 (Web Text))

Max Activating Token Index: 980

Text #5

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5293. Min Act: -0.0001

Data Index: 148277 (C4 (Web Text))

Max Activating Token Index: 827

Text #6

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.4913. Min Act: -0.0002

Data Index: 323751 (C4 (Web Text))

Max Activating Token Index: 593

Text #7

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5516. Min Act: -0.0002

Data Index: 771286 (C4 (Web Text))

Max Activating Token Index: 787

Text #8

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5047. Min Act: -0.0001

Data Index: 227890 (C4 (Web Text))

Max Activating Token Index: 813

Text #9

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5813. Min Act: -0.0002

Data Index: 987795 (C4 (Web Text))

Max Activating Token Index: 861

Text #10

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5123. Min Act: -0.0002

Data Index: 1204156 (C4 (Web Text))

Max Activating Token Index: 260

Text #11

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5103. Min Act: -0.0001

Data Index: 90498 (C4 (Web Text))

Max Activating Token Index: 151

Text #12

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5532. Min Act: -0.0001

Data Index: 469073 (C4 (Web Text))

Max Activating Token Index: 259

Text #13

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5338. Min Act: -0.0002

Data Index: 1085033 (C4 (Web Text))

Max Activating Token Index: 614

Text #14

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5658. Min Act: -0.0002

Data Index: 1262735 (C4 (Web Text))

Max Activating Token Index: 921

Text #15

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5130. Min Act: -0.0001

Data Index: 259913 (C4 (Web Text))

Max Activating Token Index: 727

Text #16

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5479. Min Act: -0.0001

Data Index: 172262 (C4 (Web Text))

Max Activating Token Index: 792

Text #17

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5752. Min Act: -0.0002

Data Index: 780963 (C4 (Web Text))

Max Activating Token Index: 1000

Text #18

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.4973. Min Act: -0.0001

Data Index: 1247587 (C4 (Web Text))

Max Activating Token Index: 696

Text #19

Max Range: 0.5859. Min Range: -0.5859

Max Act: 0.5086. Min Act: -0.0002

Data Index: 828329 (C4 (Web Text))

Max Activating Token Index: 812