Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3498 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.8067. Min Act: -0.0000

Data Index: 1366668 (Python Code)

Max Activating Token Index: 374

Text #1

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.4137. Min Act: -0.0000

Data Index: 714432 (C4 (Web Text))

Max Activating Token Index: 755

Text #2

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.5336. Min Act: -0.0000

Data Index: 1454592 (Python Code)

Max Activating Token Index: 359

Text #3

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.4284. Min Act: -0.0000

Data Index: 25968 (C4 (Web Text))

Max Activating Token Index: 24

Text #4

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.4268. Min Act: -0.0000

Data Index: 1377913 (Python Code)

Max Activating Token Index: 428

Text #5

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.1767. Min Act: -0.0000

Data Index: 527995 (C4 (Web Text))

Max Activating Token Index: 10

Text #6

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.0777. Min Act: -0.0000

Data Index: 1326838 (C4 (Web Text))

Max Activating Token Index: 991

Text #7

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.8214. Min Act: -0.0000

Data Index: 1436169 (Python Code)

Max Activating Token Index: 229

Text #8

Max Range: 2.8214. Min Range: -2.8214

Max Act: 1.9466. Min Act: -0.0000

Data Index: 1482914 (Python Code)

Max Activating Token Index: 722

Text #9

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.2448. Min Act: -0.0000

Data Index: 948234 (C4 (Web Text))

Max Activating Token Index: 147

Text #10

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.1998. Min Act: -0.0000

Data Index: 551277 (C4 (Web Text))

Max Activating Token Index: 22

Text #11

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.0518. Min Act: -0.0000

Data Index: 597172 (C4 (Web Text))

Max Activating Token Index: 871

Text #12

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.2686. Min Act: -0.0000

Data Index: 1362759 (Python Code)

Max Activating Token Index: 57

Text #13

Max Range: 2.8214. Min Range: -2.8214

Max Act: 1.9672. Min Act: -0.0000

Data Index: 441933 (C4 (Web Text))

Max Activating Token Index: 6

Text #14

Max Range: 2.8214. Min Range: -2.8214

Max Act: 1.9558. Min Act: -0.0000

Data Index: 558659 (C4 (Web Text))

Max Activating Token Index: 83

Text #15

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.1892. Min Act: -0.0000

Data Index: 1557216 (Python Code)

Max Activating Token Index: 559

Text #16

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.5789. Min Act: -0.0000

Data Index: 1426801 (Python Code)

Max Activating Token Index: 364

Text #17

Max Range: 2.8214. Min Range: -2.8214

Max Act: 2.1511. Min Act: -0.0000

Data Index: 50369 (C4 (Web Text))

Max Activating Token Index: 597

Text #18

Max Range: 2.8214. Min Range: -2.8214

Max Act: 1.9980. Min Act: -0.0000

Data Index: 1525472 (Python Code)

Max Activating Token Index: 310

Text #19

Max Range: 2.8214. Min Range: -2.8214

Max Act: 1.9999. Min Act: -0.0000

Data Index: 791659 (C4 (Web Text))

Max Activating Token Index: 694