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

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

Neuron 3686 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: 0.9944. Min Range: -0.9944

Max Act: 0.9944. Min Act: -0.0000

Data Index: 501735 (C4 (Web Text))

Max Activating Token Index: 247

Text #1

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.8975. Min Act: -0.0000

Data Index: 422940 (C4 (Web Text))

Max Activating Token Index: 946

Text #2

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.8838. Min Act: -0.0000

Data Index: 458475 (C4 (Web Text))

Max Activating Token Index: 266

Text #3

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.8257. Min Act: -0.0000

Data Index: 1219162 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.8213. Min Act: -0.0000

Data Index: 786341 (C4 (Web Text))

Max Activating Token Index: 105

Text #5

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.8010. Min Act: -0.0000

Data Index: 948412 (C4 (Web Text))

Max Activating Token Index: 796

Text #6

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7986. Min Act: -0.0000

Data Index: 709864 (C4 (Web Text))

Max Activating Token Index: 631

Text #7

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7383. Min Act: -0.0000

Data Index: 516192 (C4 (Web Text))

Max Activating Token Index: 992

Text #8

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7341. Min Act: -0.0000

Data Index: 1352035 (C4 (Web Text))

Max Activating Token Index: 312

Text #9

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7222. Min Act: -0.0000

Data Index: 477134 (C4 (Web Text))

Max Activating Token Index: 685

Text #10

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7184. Min Act: -0.0000

Data Index: 125667 (C4 (Web Text))

Max Activating Token Index: 692

Text #11

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7368. Min Act: -0.0000

Data Index: 51613 (C4 (Web Text))

Max Activating Token Index: 242

Text #12

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.6919. Min Act: -0.0000

Data Index: 1437320 (Python Code)

Max Activating Token Index: 532

Text #13

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.6927. Min Act: -0.0000

Data Index: 274459 (C4 (Web Text))

Max Activating Token Index: 748

Text #14

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7045. Min Act: -0.0000

Data Index: 1133714 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7056. Min Act: -0.0000

Data Index: 351684 (C4 (Web Text))

Max Activating Token Index: 68

Text #16

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.6765. Min Act: -0.0000

Data Index: 431124 (C4 (Web Text))

Max Activating Token Index: 942

Text #17

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.7200. Min Act: -0.0000

Data Index: 174661 (C4 (Web Text))

Max Activating Token Index: 189

Text #18

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.6957. Min Act: -0.0000

Data Index: 977909 (C4 (Web Text))

Max Activating Token Index: 607

Text #19

Max Range: 0.9944. Min Range: -0.9944

Max Act: 0.6645. Min Act: -0.0000

Data Index: 284295 (C4 (Web Text))

Max Activating Token Index: 221