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

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

Neuron 4412 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: 1.3833. Min Range: -1.3833

Max Act: 1.3833. Min Act: -0.0000

Data Index: 1643024 (Python Code)

Max Activating Token Index: 393

Text #1

Max Range: 1.3833. Min Range: -1.3833

Max Act: 1.2973. Min Act: -0.0000

Data Index: 1553332 (Python Code)

Max Activating Token Index: 188

Text #2

Max Range: 1.3833. Min Range: -1.3833

Max Act: 1.1893. Min Act: -0.0000

Data Index: 1531093 (Python Code)

Max Activating Token Index: 324

Text #3

Max Range: 1.3833. Min Range: -1.3833

Max Act: 1.2618. Min Act: -0.0000

Data Index: 1620986 (Python Code)

Max Activating Token Index: 137

Text #4

Max Range: 1.3833. Min Range: -1.3833

Max Act: 1.1597. Min Act: -0.0000

Data Index: 1409190 (Python Code)

Max Activating Token Index: 93

Text #5

Max Range: 1.3833. Min Range: -1.3833

Max Act: 1.0273. Min Act: -0.0000

Data Index: 1393852 (Python Code)

Max Activating Token Index: 89

Text #6

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.8873. Min Act: -0.0000

Data Index: 921290 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.8335. Min Act: -0.0000

Data Index: 117600 (C4 (Web Text))

Max Activating Token Index: 392

Text #8

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7743. Min Act: -0.0000

Data Index: 1039142 (C4 (Web Text))

Max Activating Token Index: 285

Text #9

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.8366. Min Act: -0.0000

Data Index: 988208 (C4 (Web Text))

Max Activating Token Index: 913

Text #10

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7639. Min Act: -0.0000

Data Index: 771929 (C4 (Web Text))

Max Activating Token Index: 784

Text #11

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.6721. Min Act: -0.0000

Data Index: 1226799 (C4 (Web Text))

Max Activating Token Index: 911

Text #12

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7310. Min Act: -0.0000

Data Index: 1635657 (Python Code)

Max Activating Token Index: 885

Text #13

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7490. Min Act: -0.0001

Data Index: 1428903 (Python Code)

Max Activating Token Index: 673

Text #14

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7352. Min Act: -0.0000

Data Index: 669817 (C4 (Web Text))

Max Activating Token Index: 293

Text #15

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7048. Min Act: -0.0000

Data Index: 90036 (C4 (Web Text))

Max Activating Token Index: 985

Text #16

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.7344. Min Act: -0.0000

Data Index: 945560 (C4 (Web Text))

Max Activating Token Index: 114

Text #17

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.6530. Min Act: -0.0000

Data Index: 440468 (C4 (Web Text))

Max Activating Token Index: 667

Text #18

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.6633. Min Act: -0.0000

Data Index: 255028 (C4 (Web Text))

Max Activating Token Index: 606

Text #19

Max Range: 1.3833. Min Range: -1.3833

Max Act: 0.6014. Min Act: -0.0000

Data Index: 1091371 (C4 (Web Text))

Max Activating Token Index: 625