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

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

Neuron 1383 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3369. Min Range: -5.3369

Max Act: 5.3369. Min Act: -0.0001

Data Index: 1272495 (C4 (Web Text))

Max Activating Token Index: 72

Text #1

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.8704. Min Act: -0.0001

Data Index: 25605 (C4 (Web Text))

Max Activating Token Index: 461

Text #2

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.7007. Min Act: -0.0000

Data Index: 1365615 (Python Code)

Max Activating Token Index: 75

Text #3

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.4090. Min Act: -0.0001

Data Index: 408746 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.4328. Min Act: -0.0001

Data Index: 636290 (C4 (Web Text))

Max Activating Token Index: 435

Text #5

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.1357. Min Act: -0.0001

Data Index: 303236 (C4 (Web Text))

Max Activating Token Index: 505

Text #6

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.2745. Min Act: -0.0001

Data Index: 133784 (C4 (Web Text))

Max Activating Token Index: 288

Text #7

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.0315. Min Act: -0.0001

Data Index: 259409 (C4 (Web Text))

Max Activating Token Index: 828

Text #8

Max Range: 5.3369. Min Range: -5.3369

Max Act: 3.1737. Min Act: -0.0000

Data Index: 1401817 (Python Code)

Max Activating Token Index: 896

Text #9

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.7133. Min Act: -0.0001

Data Index: 235140 (C4 (Web Text))

Max Activating Token Index: 38

Text #10

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.9570. Min Act: -0.0001

Data Index: 723844 (C4 (Web Text))

Max Activating Token Index: 317

Text #11

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.7952. Min Act: -0.0001

Data Index: 868837 (C4 (Web Text))

Max Activating Token Index: 806

Text #12

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.8803. Min Act: -0.0001

Data Index: 900396 (C4 (Web Text))

Max Activating Token Index: 793

Text #13

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.5829. Min Act: -0.0001

Data Index: 364500 (C4 (Web Text))

Max Activating Token Index: 155

Text #14

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.6917. Min Act: -0.0001

Data Index: 1114247 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.4818. Min Act: -0.0000

Data Index: 1549289 (Python Code)

Max Activating Token Index: 648

Text #16

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.5473. Min Act: -0.0001

Data Index: 941985 (C4 (Web Text))

Max Activating Token Index: 262

Text #17

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.4562. Min Act: -0.0001

Data Index: 657890 (C4 (Web Text))

Max Activating Token Index: 145

Text #18

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.4354. Min Act: -0.0001

Data Index: 434472 (C4 (Web Text))

Max Activating Token Index: 962

Text #19

Max Range: 5.3369. Min Range: -5.3369

Max Act: 2.2997. Min Act: -0.0001

Data Index: 724411 (C4 (Web Text))

Max Activating Token Index: 674