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

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

Neuron 3091 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1217. Min Range: -1.1217

Max Act: 1.0615. Min Act: -0.0000

Data Index: 1567665 (Python Code)

Max Activating Token Index: 414

Text #1

Max Range: 1.1217. Min Range: -1.1217

Max Act: 1.1217. Min Act: -0.0000

Data Index: 1458554 (Python Code)

Max Activating Token Index: 910

Text #2

Max Range: 1.1217. Min Range: -1.1217

Max Act: 1.0571. Min Act: -0.0000

Data Index: 1574401 (Python Code)

Max Activating Token Index: 896

Text #3

Max Range: 1.1217. Min Range: -1.1217

Max Act: 1.0196. Min Act: -0.0000

Data Index: 671511 (C4 (Web Text))

Max Activating Token Index: 281

Text #4

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.9427. Min Act: -0.0000

Data Index: 1509271 (Python Code)

Max Activating Token Index: 308

Text #5

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8796. Min Act: -0.0000

Data Index: 1362198 (Python Code)

Max Activating Token Index: 740

Text #6

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8674. Min Act: -0.0000

Data Index: 1118389 (C4 (Web Text))

Max Activating Token Index: 1021

Text #7

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8718. Min Act: -0.0000

Data Index: 1265114 (C4 (Web Text))

Max Activating Token Index: 19

Text #8

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8729. Min Act: -0.0000

Data Index: 1650262 (Python Code)

Max Activating Token Index: 895

Text #9

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8536. Min Act: -0.0000

Data Index: 1414763 (Python Code)

Max Activating Token Index: 640

Text #10

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.8189. Min Act: -0.0000

Data Index: 115126 (C4 (Web Text))

Max Activating Token Index: 900

Text #11

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7208. Min Act: -0.0000

Data Index: 1606493 (Python Code)

Max Activating Token Index: 619

Text #12

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7408. Min Act: -0.0000

Data Index: 1482042 (Python Code)

Max Activating Token Index: 883

Text #13

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7040. Min Act: -0.0000

Data Index: 9439 (C4 (Web Text))

Max Activating Token Index: 515

Text #14

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7150. Min Act: -0.0000

Data Index: 1431078 (Python Code)

Max Activating Token Index: 1006

Text #15

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7135. Min Act: -0.0000

Data Index: 17986 (C4 (Web Text))

Max Activating Token Index: 90

Text #16

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.6885. Min Act: -0.0000

Data Index: 119853 (C4 (Web Text))

Max Activating Token Index: 615

Text #17

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.6887. Min Act: -0.0000

Data Index: 1368015 (Python Code)

Max Activating Token Index: 443

Text #18

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.7123. Min Act: -0.0000

Data Index: 1389565 (Python Code)

Max Activating Token Index: 850

Text #19

Max Range: 1.1217. Min Range: -1.1217

Max Act: 0.6645. Min Act: -0.0000

Data Index: 1496874 (Python Code)

Max Activating Token Index: 860