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

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

Neuron 5571 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: 2.8172. Min Range: -2.8172

Max Act: 2.8172. Min Act: -0.0001

Data Index: 1459687 (Python Code)

Max Activating Token Index: 829

Text #1

Max Range: 2.8172. Min Range: -2.8172

Max Act: 2.0887. Min Act: -0.0000

Data Index: 1196998 (C4 (Web Text))

Max Activating Token Index: 673

Text #2

Max Range: 2.8172. Min Range: -2.8172

Max Act: 2.1623. Min Act: -0.0000

Data Index: 904734 (C4 (Web Text))

Max Activating Token Index: 24

Text #3

Max Range: 2.8172. Min Range: -2.8172

Max Act: 2.1786. Min Act: -0.0001

Data Index: 1005877 (C4 (Web Text))

Max Activating Token Index: 474

Text #4

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.9483. Min Act: -0.0001

Data Index: 1147645 (C4 (Web Text))

Max Activating Token Index: 107

Text #5

Max Range: 2.8172. Min Range: -2.8172

Max Act: 2.1418. Min Act: -0.0001

Data Index: 852538 (C4 (Web Text))

Max Activating Token Index: 5

Text #6

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.9999. Min Act: -0.0000

Data Index: 701106 (C4 (Web Text))

Max Activating Token Index: 331

Text #7

Max Range: 2.8172. Min Range: -2.8172

Max Act: 2.0758. Min Act: -0.0000

Data Index: 374748 (C4 (Web Text))

Max Activating Token Index: 482

Text #8

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.8615. Min Act: -0.0000

Data Index: 509740 (C4 (Web Text))

Max Activating Token Index: 236

Text #9

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.9614. Min Act: -0.0000

Data Index: 386525 (C4 (Web Text))

Max Activating Token Index: 13

Text #10

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.9807. Min Act: -0.0000

Data Index: 508959 (C4 (Web Text))

Max Activating Token Index: 22

Text #11

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7464. Min Act: -0.0001

Data Index: 183835 (C4 (Web Text))

Max Activating Token Index: 192

Text #12

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7600. Min Act: -0.0001

Data Index: 805160 (C4 (Web Text))

Max Activating Token Index: 7

Text #13

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7155. Min Act: -0.0000

Data Index: 284431 (C4 (Web Text))

Max Activating Token Index: 777

Text #14

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.5587. Min Act: -0.0000

Data Index: 825559 (C4 (Web Text))

Max Activating Token Index: 882

Text #15

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7456. Min Act: -0.0000

Data Index: 802145 (C4 (Web Text))

Max Activating Token Index: 258

Text #16

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.8001. Min Act: -0.0000

Data Index: 1469957 (Python Code)

Max Activating Token Index: 33

Text #17

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.6802. Min Act: -0.0000

Data Index: 214348 (C4 (Web Text))

Max Activating Token Index: 610

Text #18

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7721. Min Act: -0.0001

Data Index: 1103093 (C4 (Web Text))

Max Activating Token Index: 616

Text #19

Max Range: 2.8172. Min Range: -2.8172

Max Act: 1.7795. Min Act: -0.0001

Data Index: 30174 (C4 (Web Text))

Max Activating Token Index: 561