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

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

Neuron 782 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: 5.3702. Min Range: -5.3702

Max Act: 5.3702. Min Act: -0.0001

Data Index: 1496272 (Python Code)

Max Activating Token Index: 774

Text #1

Max Range: 5.3702. Min Range: -5.3702

Max Act: 5.1366. Min Act: -0.0001

Data Index: 1562084 (Python Code)

Max Activating Token Index: 187

Text #2

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.7792. Min Act: -0.0001

Data Index: 1639912 (Python Code)

Max Activating Token Index: 169

Text #3

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.7988. Min Act: -0.0001

Data Index: 1561785 (Python Code)

Max Activating Token Index: 561

Text #4

Max Range: 5.3702. Min Range: -5.3702

Max Act: 5.1130. Min Act: -0.0001

Data Index: 1364128 (Python Code)

Max Activating Token Index: 355

Text #5

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.9505. Min Act: -0.0001

Data Index: 1554629 (Python Code)

Max Activating Token Index: 323

Text #6

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.6712. Min Act: -0.0001

Data Index: 1361367 (Python Code)

Max Activating Token Index: 645

Text #7

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.5935. Min Act: -0.0001

Data Index: 1540369 (Python Code)

Max Activating Token Index: 35

Text #8

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.6713. Min Act: -0.0001

Data Index: 1564623 (Python Code)

Max Activating Token Index: 271

Text #9

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.4934. Min Act: -0.0001

Data Index: 1608807 (Python Code)

Max Activating Token Index: 622

Text #10

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.4369. Min Act: -0.0001

Data Index: 1586345 (Python Code)

Max Activating Token Index: 209

Text #11

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.1932. Min Act: -0.0001

Data Index: 1562435 (Python Code)

Max Activating Token Index: 174

Text #12

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.1515. Min Act: -0.0001

Data Index: 1633609 (Python Code)

Max Activating Token Index: 352

Text #13

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.3356. Min Act: -0.0001

Data Index: 1495292 (Python Code)

Max Activating Token Index: 439

Text #14

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.2845. Min Act: -0.0001

Data Index: 1448229 (Python Code)

Max Activating Token Index: 825

Text #15

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.1227. Min Act: -0.0001

Data Index: 1520762 (Python Code)

Max Activating Token Index: 9

Text #16

Max Range: 5.3702. Min Range: -5.3702

Max Act: 3.9414. Min Act: -0.0001

Data Index: 1508395 (Python Code)

Max Activating Token Index: 124

Text #17

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.2536. Min Act: -0.0001

Data Index: 1503869 (Python Code)

Max Activating Token Index: 513

Text #18

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.1892. Min Act: -0.0001

Data Index: 1431241 (Python Code)

Max Activating Token Index: 665

Text #19

Max Range: 5.3702. Min Range: -5.3702

Max Act: 4.0065. Min Act: -0.0001

Data Index: 1652048 (Python Code)

Max Activating Token Index: 161