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

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

Neuron 3645 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: 2.8012. Min Range: -2.8012

Max Act: 2.8011. Min Act: -0.0000

Data Index: 1281641 (C4 (Web Text))

Max Activating Token Index: 380

Text #1

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.5807. Min Act: -0.0000

Data Index: 1326755 (C4 (Web Text))

Max Activating Token Index: 991

Text #2

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.8012. Min Act: -0.0000

Data Index: 1233727 (C4 (Web Text))

Max Activating Token Index: 302

Text #3

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.6661. Min Act: -0.0000

Data Index: 1232620 (C4 (Web Text))

Max Activating Token Index: 281

Text #4

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.7820. Min Act: -0.0000

Data Index: 782754 (C4 (Web Text))

Max Activating Token Index: 233

Text #5

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.5226. Min Act: -0.0000

Data Index: 58339 (C4 (Web Text))

Max Activating Token Index: 746

Text #6

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.4945. Min Act: -0.0000

Data Index: 368008 (C4 (Web Text))

Max Activating Token Index: 325

Text #7

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.4606. Min Act: -0.0000

Data Index: 1234132 (C4 (Web Text))

Max Activating Token Index: 770

Text #8

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.4097. Min Act: -0.0000

Data Index: 667576 (C4 (Web Text))

Max Activating Token Index: 406

Text #9

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.3814. Min Act: -0.0000

Data Index: 158097 (C4 (Web Text))

Max Activating Token Index: 156

Text #10

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.2436. Min Act: -0.0000

Data Index: 1184876 (C4 (Web Text))

Max Activating Token Index: 411

Text #11

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.3466. Min Act: -0.0000

Data Index: 572480 (C4 (Web Text))

Max Activating Token Index: 816

Text #12

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.3391. Min Act: -0.0000

Data Index: 983689 (C4 (Web Text))

Max Activating Token Index: 938

Text #13

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.2916. Min Act: -0.0000

Data Index: 991289 (C4 (Web Text))

Max Activating Token Index: 146

Text #14

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.3960. Min Act: -0.0000

Data Index: 1275535 (C4 (Web Text))

Max Activating Token Index: 870

Text #15

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.1736. Min Act: -0.0000

Data Index: 1309602 (C4 (Web Text))

Max Activating Token Index: 875

Text #16

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.1676. Min Act: -0.0000

Data Index: 1570700 (Python Code)

Max Activating Token Index: 173

Text #17

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.1610. Min Act: -0.0000

Data Index: 682714 (C4 (Web Text))

Max Activating Token Index: 360

Text #18

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.1537. Min Act: -0.0000

Data Index: 435183 (C4 (Web Text))

Max Activating Token Index: 433

Text #19

Max Range: 2.8012. Min Range: -2.8012

Max Act: 2.3892. Min Act: -0.0000

Data Index: 789957 (C4 (Web Text))

Max Activating Token Index: 262