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

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

Neuron 2604 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.9979. Min Range: -8.9979

Max Act: 8.9979. Min Act: -0.0000

Data Index: 861958 (C4 (Web Text))

Max Activating Token Index: 202

Text #1

Max Range: 8.9979. Min Range: -8.9979

Max Act: 8.2347. Min Act: -0.0000

Data Index: 426794 (C4 (Web Text))

Max Activating Token Index: 496

Text #2

Max Range: 8.9979. Min Range: -8.9979

Max Act: 7.3835. Min Act: -0.0000

Data Index: 890063 (C4 (Web Text))

Max Activating Token Index: 834

Text #3

Max Range: 8.9979. Min Range: -8.9979

Max Act: 6.8991. Min Act: -0.0000

Data Index: 761712 (C4 (Web Text))

Max Activating Token Index: 299

Text #4

Max Range: 8.9979. Min Range: -8.9979

Max Act: 6.8720. Min Act: -0.0000

Data Index: 189363 (C4 (Web Text))

Max Activating Token Index: 453

Text #5

Max Range: 8.9979. Min Range: -8.9979

Max Act: 6.8908. Min Act: -0.0000

Data Index: 1221614 (C4 (Web Text))

Max Activating Token Index: 536

Text #6

Max Range: 8.9979. Min Range: -8.9979

Max Act: 5.8704. Min Act: -0.0000

Data Index: 220473 (C4 (Web Text))

Max Activating Token Index: 141

Text #7

Max Range: 8.9979. Min Range: -8.9979

Max Act: 5.6843. Min Act: -0.0000

Data Index: 52919 (C4 (Web Text))

Max Activating Token Index: 255

Text #8

Max Range: 8.9979. Min Range: -8.9979

Max Act: 4.8558. Min Act: -0.0000

Data Index: 648997 (C4 (Web Text))

Max Activating Token Index: 820

Text #9

Max Range: 8.9979. Min Range: -8.9979

Max Act: 4.6398. Min Act: -0.0000

Data Index: 939096 (C4 (Web Text))

Max Activating Token Index: 947

Text #10

Max Range: 8.9979. Min Range: -8.9979

Max Act: 4.8258. Min Act: -0.0000

Data Index: 1112149 (C4 (Web Text))

Max Activating Token Index: 804

Text #11

Max Range: 8.9979. Min Range: -8.9979

Max Act: 5.1453. Min Act: -0.0000

Data Index: 748148 (C4 (Web Text))

Max Activating Token Index: 837

Text #12

Max Range: 8.9979. Min Range: -8.9979

Max Act: 3.0050. Min Act: -0.0000

Data Index: 826133 (C4 (Web Text))

Max Activating Token Index: 826

Text #13

Max Range: 8.9979. Min Range: -8.9979

Max Act: 2.3813. Min Act: -0.0000

Data Index: 665336 (C4 (Web Text))

Max Activating Token Index: 984

Text #14

Max Range: 8.9979. Min Range: -8.9979

Max Act: 4.0214. Min Act: -0.0000

Data Index: 1338892 (C4 (Web Text))

Max Activating Token Index: 243

Text #15

Max Range: 8.9979. Min Range: -8.9979

Max Act: 3.1695. Min Act: -0.0000

Data Index: 336986 (C4 (Web Text))

Max Activating Token Index: 974

Text #16

Max Range: 8.9979. Min Range: -8.9979

Max Act: 2.8953. Min Act: -0.0000

Data Index: 736631 (C4 (Web Text))

Max Activating Token Index: 39

Text #17

Max Range: 8.9979. Min Range: -8.9979

Max Act: 2.4756. Min Act: -0.0000

Data Index: 839093 (C4 (Web Text))

Max Activating Token Index: 568

Text #18

Max Range: 8.9979. Min Range: -8.9979

Max Act: 2.5095. Min Act: -0.0000

Data Index: 843776 (C4 (Web Text))

Max Activating Token Index: 60

Text #19

Max Range: 8.9979. Min Range: -8.9979

Max Act: 2.5079. Min Act: -0.0000

Data Index: 1236677 (C4 (Web Text))

Max Activating Token Index: 555