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

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

Neuron 2465 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: 0.8105. Min Range: -0.8105

Max Act: 0.8105. Min Act: -0.0000

Data Index: 1388235 (Python Code)

Max Activating Token Index: 698

Text #1

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.6122. Min Act: -0.0001

Data Index: 1544704 (Python Code)

Max Activating Token Index: 313

Text #2

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.5358. Min Act: -0.0000

Data Index: 585624 (C4 (Web Text))

Max Activating Token Index: 878

Text #3

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.5500. Min Act: -0.0000

Data Index: 1214875 (C4 (Web Text))

Max Activating Token Index: 934

Text #4

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4770. Min Act: -0.0000

Data Index: 1537967 (Python Code)

Max Activating Token Index: 393

Text #5

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4761. Min Act: -0.0001

Data Index: 854041 (C4 (Web Text))

Max Activating Token Index: 531

Text #6

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.5346. Min Act: -0.0000

Data Index: 1459001 (Python Code)

Max Activating Token Index: 512

Text #7

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.5252. Min Act: -0.0000

Data Index: 1409593 (Python Code)

Max Activating Token Index: 60

Text #8

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4489. Min Act: -0.0000

Data Index: 443341 (C4 (Web Text))

Max Activating Token Index: 730

Text #9

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4041. Min Act: -0.0000

Data Index: 981651 (C4 (Web Text))

Max Activating Token Index: 140

Text #10

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4621. Min Act: -0.0000

Data Index: 1274925 (C4 (Web Text))

Max Activating Token Index: 89

Text #11

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4582. Min Act: -0.0000

Data Index: 1561333 (Python Code)

Max Activating Token Index: 26

Text #12

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4706. Min Act: -0.0000

Data Index: 351734 (C4 (Web Text))

Max Activating Token Index: 361

Text #13

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4354. Min Act: -0.0000

Data Index: 1501426 (Python Code)

Max Activating Token Index: 716

Text #14

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4517. Min Act: -0.0000

Data Index: 1145314 (C4 (Web Text))

Max Activating Token Index: 439

Text #15

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4663. Min Act: -0.0000

Data Index: 1444869 (Python Code)

Max Activating Token Index: 669

Text #16

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4997. Min Act: -0.0000

Data Index: 1208338 (C4 (Web Text))

Max Activating Token Index: 740

Text #17

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4341. Min Act: -0.0000

Data Index: 1434427 (Python Code)

Max Activating Token Index: 805

Text #18

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4387. Min Act: -0.0000

Data Index: 1081929 (C4 (Web Text))

Max Activating Token Index: 385

Text #19

Max Range: 0.8105. Min Range: -0.8105

Max Act: 0.4081. Min Act: -0.0000

Data Index: 959214 (C4 (Web Text))

Max Activating Token Index: 588