Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2957 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.6984. Min Act: -0.0001

Data Index: 1524292 (Python Code)

Max Activating Token Index: 692

Text #1

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.5818. Min Act: -0.0001

Data Index: 1414841 (Python Code)

Max Activating Token Index: 165

Text #2

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4997. Min Act: -0.0000

Data Index: 1151460 (C4 (Web Text))

Max Activating Token Index: 614

Text #3

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.5035. Min Act: -0.0001

Data Index: 963138 (C4 (Web Text))

Max Activating Token Index: 668

Text #4

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4945. Min Act: -0.0001

Data Index: 1173320 (C4 (Web Text))

Max Activating Token Index: 500

Text #5

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4805. Min Act: -0.0000

Data Index: 862202 (C4 (Web Text))

Max Activating Token Index: 139

Text #6

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4828. Min Act: -0.0000

Data Index: 129490 (C4 (Web Text))

Max Activating Token Index: 282

Text #7

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4608. Min Act: -0.0001

Data Index: 1038544 (C4 (Web Text))

Max Activating Token Index: 46

Text #8

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4250. Min Act: -0.0001

Data Index: 100357 (C4 (Web Text))

Max Activating Token Index: 612

Text #9

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4474. Min Act: -0.0001

Data Index: 1390955 (Python Code)

Max Activating Token Index: 14

Text #10

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4341. Min Act: -0.0001

Data Index: 1572896 (Python Code)

Max Activating Token Index: 150

Text #11

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4136. Min Act: -0.0001

Data Index: 427402 (C4 (Web Text))

Max Activating Token Index: 26

Text #12

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4037. Min Act: -0.0001

Data Index: 1578832 (Python Code)

Max Activating Token Index: 196

Text #13

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4660. Min Act: -0.0000

Data Index: 793712 (C4 (Web Text))

Max Activating Token Index: 30

Text #14

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.3968. Min Act: -0.0001

Data Index: 1413352 (Python Code)

Max Activating Token Index: 769

Text #15

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4049. Min Act: -0.0001

Data Index: 1405842 (Python Code)

Max Activating Token Index: 101

Text #16

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.3808. Min Act: -0.0001

Data Index: 1179153 (C4 (Web Text))

Max Activating Token Index: 1003

Text #17

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.4009. Min Act: -0.0001

Data Index: 1285925 (C4 (Web Text))

Max Activating Token Index: 65

Text #18

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.3748. Min Act: -0.0001

Data Index: 1372663 (Python Code)

Max Activating Token Index: 781

Text #19

Max Range: 0.6984. Min Range: -0.6984

Max Act: 0.3965. Min Act: -0.0001

Data Index: 1649845 (Python Code)

Max Activating Token Index: 37