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

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

Neuron 3900 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1523. Min Range: -5.1523

Max Act: 4.6428. Min Act: -0.0002

Data Index: 415505 (C4 (Web Text))

Max Activating Token Index: 399

Text #1

Max Range: 5.1523. Min Range: -5.1523

Max Act: 5.1523. Min Act: -0.0002

Data Index: 1121917 (C4 (Web Text))

Max Activating Token Index: 901

Text #2

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.9948. Min Act: -0.0002

Data Index: 879636 (C4 (Web Text))

Max Activating Token Index: 442

Text #3

Max Range: 5.1523. Min Range: -5.1523

Max Act: 3.7717. Min Act: -0.0002

Data Index: 1004126 (C4 (Web Text))

Max Activating Token Index: 635

Text #4

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.8015. Min Act: -0.0002

Data Index: 283812 (C4 (Web Text))

Max Activating Token Index: 44

Text #5

Max Range: 5.1523. Min Range: -5.1523

Max Act: 3.0797. Min Act: -0.0002

Data Index: 914223 (C4 (Web Text))

Max Activating Token Index: 67

Text #6

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.7248. Min Act: -0.0002

Data Index: 854916 (C4 (Web Text))

Max Activating Token Index: 156

Text #7

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.6004. Min Act: -0.0002

Data Index: 382629 (C4 (Web Text))

Max Activating Token Index: 915

Text #8

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.5993. Min Act: -0.0002

Data Index: 480200 (C4 (Web Text))

Max Activating Token Index: 604

Text #9

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.8427. Min Act: -0.0002

Data Index: 43198 (C4 (Web Text))

Max Activating Token Index: 274

Text #10

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.7388. Min Act: -0.0002

Data Index: 1045722 (C4 (Web Text))

Max Activating Token Index: 245

Text #11

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.4853. Min Act: -0.0002

Data Index: 19356 (C4 (Web Text))

Max Activating Token Index: 139

Text #12

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.6564. Min Act: -0.0002

Data Index: 1320595 (C4 (Web Text))

Max Activating Token Index: 247

Text #13

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.5551. Min Act: -0.0002

Data Index: 326568 (C4 (Web Text))

Max Activating Token Index: 198

Text #14

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.7237. Min Act: -0.0002

Data Index: 741160 (C4 (Web Text))

Max Activating Token Index: 138

Text #15

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.3030. Min Act: -0.0002

Data Index: 1033802 (C4 (Web Text))

Max Activating Token Index: 622

Text #16

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.5430. Min Act: -0.0002

Data Index: 1299094 (C4 (Web Text))

Max Activating Token Index: 447

Text #17

Max Range: 5.1523. Min Range: -5.1523

Max Act: 3.3145. Min Act: -0.0002

Data Index: 758137 (C4 (Web Text))

Max Activating Token Index: 799

Text #18

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.5922. Min Act: -0.0002

Data Index: 764817 (C4 (Web Text))

Max Activating Token Index: 227

Text #19

Max Range: 5.1523. Min Range: -5.1523

Max Act: 2.3554. Min Act: -0.0002

Data Index: 853231 (C4 (Web Text))

Max Activating Token Index: 54