Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3703 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.5581. Min Act: -0.0002

Data Index: 1540337 (Python Code)

Max Activating Token Index: 838

Text #1

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.6113. Min Act: -0.0002

Data Index: 1590963 (Python Code)

Max Activating Token Index: 845

Text #2

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.6241. Min Act: -0.0002

Data Index: 1545290 (Python Code)

Max Activating Token Index: 742

Text #3

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4913. Min Act: -0.0002

Data Index: 1544367 (Python Code)

Max Activating Token Index: 731

Text #4

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4853. Min Act: -0.0002

Data Index: 1588440 (Python Code)

Max Activating Token Index: 588

Text #5

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4798. Min Act: -0.0002

Data Index: 1490953 (Python Code)

Max Activating Token Index: 854

Text #6

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4203. Min Act: -0.0002

Data Index: 1520609 (Python Code)

Max Activating Token Index: 765

Text #7

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4964. Min Act: -0.0002

Data Index: 1618807 (Python Code)

Max Activating Token Index: 757

Text #8

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4022. Min Act: -0.0002

Data Index: 1485838 (Python Code)

Max Activating Token Index: 845

Text #9

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3780. Min Act: -0.0002

Data Index: 1563886 (Python Code)

Max Activating Token Index: 867

Text #10

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4614. Min Act: -0.0002

Data Index: 1511927 (Python Code)

Max Activating Token Index: 552

Text #11

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3929. Min Act: -0.0002

Data Index: 1369513 (Python Code)

Max Activating Token Index: 954

Text #12

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3619. Min Act: -0.0002

Data Index: 1473433 (Python Code)

Max Activating Token Index: 767

Text #13

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4119. Min Act: -0.0002

Data Index: 1544924 (Python Code)

Max Activating Token Index: 26

Text #14

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4616. Min Act: -0.0002

Data Index: 1364441 (Python Code)

Max Activating Token Index: 720

Text #15

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.4358. Min Act: -0.0002

Data Index: 1440328 (Python Code)

Max Activating Token Index: 738

Text #16

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3850. Min Act: -0.0002

Data Index: 1545639 (Python Code)

Max Activating Token Index: 188

Text #17

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3793. Min Act: -0.0002

Data Index: 1539137 (Python Code)

Max Activating Token Index: 712

Text #18

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.3619. Min Act: -0.0002

Data Index: 1607724 (Python Code)

Max Activating Token Index: 452

Text #19

Max Range: 2.6241. Min Range: -2.6241

Max Act: 2.2990. Min Act: -0.0002

Data Index: 1522069 (Python Code)

Max Activating Token Index: 572