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

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

Neuron 4862 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: 3.1751. Min Range: -3.1751

Max Act: 2.6417. Min Act: -0.0001

Data Index: 193868 (C4 (Web Text))

Max Activating Token Index: 689

Text #1

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.7526. Min Act: -0.0001

Data Index: 810615 (C4 (Web Text))

Max Activating Token Index: 87

Text #2

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.6373. Min Act: -0.0001

Data Index: 411484 (C4 (Web Text))

Max Activating Token Index: 625

Text #3

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.7273. Min Act: -0.0001

Data Index: 656667 (C4 (Web Text))

Max Activating Token Index: 228

Text #4

Max Range: 3.1751. Min Range: -3.1751

Max Act: 1.9392. Min Act: -0.0001

Data Index: 740237 (C4 (Web Text))

Max Activating Token Index: 793

Text #5

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.2279. Min Act: -0.0001

Data Index: 228990 (C4 (Web Text))

Max Activating Token Index: 156

Text #6

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.5950. Min Act: -0.0001

Data Index: 424540 (C4 (Web Text))

Max Activating Token Index: 622

Text #7

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.6924. Min Act: -0.0001

Data Index: 1208531 (C4 (Web Text))

Max Activating Token Index: 1005

Text #8

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.6633. Min Act: -0.0001

Data Index: 245047 (C4 (Web Text))

Max Activating Token Index: 907

Text #9

Max Range: 3.1751. Min Range: -3.1751

Max Act: 1.9655. Min Act: -0.0001

Data Index: 1031464 (C4 (Web Text))

Max Activating Token Index: 162

Text #10

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.5485. Min Act: -0.0001

Data Index: 1221234 (C4 (Web Text))

Max Activating Token Index: 788

Text #11

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.4265. Min Act: -0.0001

Data Index: 623467 (C4 (Web Text))

Max Activating Token Index: 907

Text #12

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.3628. Min Act: -0.0001

Data Index: 759077 (C4 (Web Text))

Max Activating Token Index: 456

Text #13

Max Range: 3.1751. Min Range: -3.1751

Max Act: 3.1751. Min Act: -0.0001

Data Index: 1100928 (C4 (Web Text))

Max Activating Token Index: 726

Text #14

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.5754. Min Act: -0.0001

Data Index: 817062 (C4 (Web Text))

Max Activating Token Index: 784

Text #15

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.0876. Min Act: -0.0001

Data Index: 1302446 (C4 (Web Text))

Max Activating Token Index: 771

Text #16

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.0916. Min Act: -0.0001

Data Index: 706879 (C4 (Web Text))

Max Activating Token Index: 208

Text #17

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.3465. Min Act: -0.0001

Data Index: 680604 (C4 (Web Text))

Max Activating Token Index: 186

Text #18

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.0742. Min Act: -0.0001

Data Index: 954131 (C4 (Web Text))

Max Activating Token Index: 865

Text #19

Max Range: 3.1751. Min Range: -3.1751

Max Act: 2.1346. Min Act: -0.0001

Data Index: 1339649 (C4 (Web Text))

Max Activating Token Index: 54