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

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

Neuron 3970 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: 1.5123. Min Range: -1.5123

Max Act: 1.5123. Min Act: -0.0000

Data Index: 1161998 (C4 (Web Text))

Max Activating Token Index: 632

Text #1

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.3500. Min Act: -0.0000

Data Index: 501852 (C4 (Web Text))

Max Activating Token Index: 755

Text #2

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.2036. Min Act: -0.0000

Data Index: 666463 (C4 (Web Text))

Max Activating Token Index: 1000

Text #3

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.2571. Min Act: -0.0000

Data Index: 846963 (C4 (Web Text))

Max Activating Token Index: 255

Text #4

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.2655. Min Act: -0.0000

Data Index: 25397 (C4 (Web Text))

Max Activating Token Index: 117

Text #5

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.1484. Min Act: -0.0000

Data Index: 1308668 (C4 (Web Text))

Max Activating Token Index: 561

Text #6

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.1539. Min Act: -0.0000

Data Index: 921222 (C4 (Web Text))

Max Activating Token Index: 383

Text #7

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.0947. Min Act: -0.0000

Data Index: 866597 (C4 (Web Text))

Max Activating Token Index: 339

Text #8

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.0690. Min Act: -0.0001

Data Index: 1481035 (Python Code)

Max Activating Token Index: 688

Text #9

Max Range: 1.5123. Min Range: -1.5123

Max Act: 1.0427. Min Act: -0.0000

Data Index: 753723 (C4 (Web Text))

Max Activating Token Index: 55

Text #10

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.9935. Min Act: -0.0000

Data Index: 781787 (C4 (Web Text))

Max Activating Token Index: 426

Text #11

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.9510. Min Act: -0.0000

Data Index: 1577328 (Python Code)

Max Activating Token Index: 866

Text #12

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.9574. Min Act: -0.0000

Data Index: 1373215 (Python Code)

Max Activating Token Index: 524

Text #13

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8571. Min Act: -0.0000

Data Index: 1428540 (Python Code)

Max Activating Token Index: 101

Text #14

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.9110. Min Act: -0.0000

Data Index: 1425304 (Python Code)

Max Activating Token Index: 44

Text #15

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8071. Min Act: -0.0000

Data Index: 1569841 (Python Code)

Max Activating Token Index: 769

Text #16

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8389. Min Act: -0.0001

Data Index: 808074 (C4 (Web Text))

Max Activating Token Index: 150

Text #17

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8231. Min Act: -0.0000

Data Index: 1540436 (Python Code)

Max Activating Token Index: 334

Text #18

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8159. Min Act: -0.0000

Data Index: 1467673 (Python Code)

Max Activating Token Index: 517

Text #19

Max Range: 1.5123. Min Range: -1.5123

Max Act: 0.8172. Min Act: -0.0000

Data Index: 1420965 (Python Code)

Max Activating Token Index: 179