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

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

Neuron 4912 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4649. Min Act: -0.0000

Data Index: 982220 (C4 (Web Text))

Max Activating Token Index: 70

Text #1

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3341. Min Act: -0.0000

Data Index: 1640418 (Python Code)

Max Activating Token Index: 791

Text #2

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4203. Min Act: -0.0000

Data Index: 921521 (C4 (Web Text))

Max Activating Token Index: 682

Text #3

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4544. Min Act: -0.0000

Data Index: 1205333 (C4 (Web Text))

Max Activating Token Index: 77

Text #4

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4188. Min Act: -0.0000

Data Index: 1562167 (Python Code)

Max Activating Token Index: 892

Text #5

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4384. Min Act: -0.0000

Data Index: 27227 (C4 (Web Text))

Max Activating Token Index: 123

Text #6

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4412. Min Act: -0.0000

Data Index: 876363 (C4 (Web Text))

Max Activating Token Index: 46

Text #7

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3754. Min Act: -0.0000

Data Index: 1414406 (Python Code)

Max Activating Token Index: 653

Text #8

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4969. Min Act: -0.0000

Data Index: 1560429 (Python Code)

Max Activating Token Index: 235

Text #9

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4072. Min Act: -0.0000

Data Index: 1198750 (C4 (Web Text))

Max Activating Token Index: 101

Text #10

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4578. Min Act: -0.0000

Data Index: 1491897 (Python Code)

Max Activating Token Index: 58

Text #11

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4234. Min Act: -0.0000

Data Index: 1265684 (C4 (Web Text))

Max Activating Token Index: 79

Text #12

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3990. Min Act: -0.0000

Data Index: 1119699 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3774. Min Act: -0.0000

Data Index: 1568834 (Python Code)

Max Activating Token Index: 945

Text #14

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3255. Min Act: -0.0000

Data Index: 147207 (C4 (Web Text))

Max Activating Token Index: 174

Text #15

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.4133. Min Act: -0.0000

Data Index: 388969 (C4 (Web Text))

Max Activating Token Index: 174

Text #16

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3798. Min Act: -0.0000

Data Index: 1437893 (Python Code)

Max Activating Token Index: 624

Text #17

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3740. Min Act: -0.0000

Data Index: 31031 (C4 (Web Text))

Max Activating Token Index: 513

Text #18

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3720. Min Act: -0.0000

Data Index: 245977 (C4 (Web Text))

Max Activating Token Index: 332

Text #19

Max Range: 0.4969. Min Range: -0.4969

Max Act: 0.3863. Min Act: -0.0000

Data Index: 1511930 (Python Code)

Max Activating Token Index: 692