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

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

Neuron 5904 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.7962. Min Act: -0.0000

Data Index: 1624163 (Python Code)

Max Activating Token Index: 167

Text #1

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.4151. Min Act: -0.0000

Data Index: 781373 (C4 (Web Text))

Max Activating Token Index: 431

Text #2

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.1553. Min Act: -0.0000

Data Index: 285320 (C4 (Web Text))

Max Activating Token Index: 511

Text #3

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.1323. Min Act: -0.0000

Data Index: 432441 (C4 (Web Text))

Max Activating Token Index: 502

Text #4

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8835. Min Act: -0.0000

Data Index: 1151554 (C4 (Web Text))

Max Activating Token Index: 410

Text #5

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.0554. Min Act: -0.0000

Data Index: 116013 (C4 (Web Text))

Max Activating Token Index: 738

Text #6

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8862. Min Act: -0.0000

Data Index: 1402940 (Python Code)

Max Activating Token Index: 244

Text #7

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.9582. Min Act: -0.0000

Data Index: 1598432 (Python Code)

Max Activating Token Index: 992

Text #8

Max Range: 2.7962. Min Range: -2.7962

Max Act: 2.0097. Min Act: -0.0000

Data Index: 799543 (C4 (Web Text))

Max Activating Token Index: 624

Text #9

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.7899. Min Act: -0.0000

Data Index: 1061548 (C4 (Web Text))

Max Activating Token Index: 38

Text #10

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.7919. Min Act: -0.0000

Data Index: 857865 (C4 (Web Text))

Max Activating Token Index: 295

Text #11

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8726. Min Act: -0.0000

Data Index: 309375 (C4 (Web Text))

Max Activating Token Index: 594

Text #12

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8737. Min Act: -0.0000

Data Index: 1438772 (Python Code)

Max Activating Token Index: 75

Text #13

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8677. Min Act: -0.0000

Data Index: 25671 (C4 (Web Text))

Max Activating Token Index: 833

Text #14

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8706. Min Act: -0.0000

Data Index: 901922 (C4 (Web Text))

Max Activating Token Index: 171

Text #15

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.9643. Min Act: -0.0000

Data Index: 1481080 (Python Code)

Max Activating Token Index: 30

Text #16

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8576. Min Act: -0.0000

Data Index: 223477 (C4 (Web Text))

Max Activating Token Index: 175

Text #17

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.7142. Min Act: -0.0000

Data Index: 1007511 (C4 (Web Text))

Max Activating Token Index: 425

Text #18

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.7944. Min Act: -0.0000

Data Index: 1056671 (C4 (Web Text))

Max Activating Token Index: 468

Text #19

Max Range: 2.7962. Min Range: -2.7962

Max Act: 1.8013. Min Act: -0.0000

Data Index: 1544280 (Python Code)

Max Activating Token Index: 415