Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 207 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.1184. Min Act: -0.0003

Data Index: 340206 (C4 (Web Text))

Max Activating Token Index: 975

Text #1

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0985. Min Act: -0.0003

Data Index: 1415326 (Python Code)

Max Activating Token Index: 397

Text #2

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0981. Min Act: -0.0003

Data Index: 502009 (C4 (Web Text))

Max Activating Token Index: 397

Text #3

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0873. Min Act: -0.0003

Data Index: 109362 (C4 (Web Text))

Max Activating Token Index: 575

Text #4

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0951. Min Act: -0.0003

Data Index: 836053 (C4 (Web Text))

Max Activating Token Index: 970

Text #5

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0941. Min Act: -0.0003

Data Index: 1322301 (C4 (Web Text))

Max Activating Token Index: 839

Text #6

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0879. Min Act: -0.0004

Data Index: 348231 (C4 (Web Text))

Max Activating Token Index: 662

Text #7

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0919. Min Act: -0.0003

Data Index: 675024 (C4 (Web Text))

Max Activating Token Index: 517

Text #8

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0826. Min Act: -0.0003

Data Index: 716710 (C4 (Web Text))

Max Activating Token Index: 190

Text #9

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0861. Min Act: -0.0003

Data Index: 429608 (C4 (Web Text))

Max Activating Token Index: 97

Text #10

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0837. Min Act: -0.0003

Data Index: 73593 (C4 (Web Text))

Max Activating Token Index: 953

Text #11

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0903. Min Act: -0.0003

Data Index: 141896 (C4 (Web Text))

Max Activating Token Index: 1022

Text #12

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0834. Min Act: -0.0003

Data Index: 1364816 (Python Code)

Max Activating Token Index: 198

Text #13

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0856. Min Act: -0.0003

Data Index: 532195 (C4 (Web Text))

Max Activating Token Index: 517

Text #14

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0852. Min Act: -0.0003

Data Index: 473123 (C4 (Web Text))

Max Activating Token Index: 506

Text #15

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0826. Min Act: -0.0002

Data Index: 819451 (C4 (Web Text))

Max Activating Token Index: 959

Text #16

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0838. Min Act: -0.0003

Data Index: 1160231 (C4 (Web Text))

Max Activating Token Index: 873

Text #17

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0804. Min Act: -0.0003

Data Index: 1335273 (C4 (Web Text))

Max Activating Token Index: 735

Text #18

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0825. Min Act: -0.0004

Data Index: 951322 (C4 (Web Text))

Max Activating Token Index: 523

Text #19

Max Range: 0.1184. Min Range: -0.1184

Max Act: 0.0837. Min Act: -0.0004

Data Index: 125507 (C4 (Web Text))

Max Activating Token Index: 226