Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1474 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.4505. Min Act: -0.0001

Data Index: 1481873 (Python Code)

Max Activating Token Index: 974

Text #1

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.4331. Min Act: -0.0000

Data Index: 882674 (C4 (Web Text))

Max Activating Token Index: 741

Text #2

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3585. Min Act: -0.0000

Data Index: 1543030 (Python Code)

Max Activating Token Index: 858

Text #3

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3364. Min Act: -0.0001

Data Index: 1626313 (Python Code)

Max Activating Token Index: 893

Text #4

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3274. Min Act: -0.0001

Data Index: 610277 (C4 (Web Text))

Max Activating Token Index: 72

Text #5

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3201. Min Act: -0.0001

Data Index: 821732 (C4 (Web Text))

Max Activating Token Index: 871

Text #6

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3242. Min Act: -0.0000

Data Index: 237296 (C4 (Web Text))

Max Activating Token Index: 600

Text #7

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2987. Min Act: -0.0000

Data Index: 563500 (C4 (Web Text))

Max Activating Token Index: 244

Text #8

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.3055. Min Act: -0.0001

Data Index: 1538909 (Python Code)

Max Activating Token Index: 663

Text #9

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2998. Min Act: -0.0000

Data Index: 414853 (C4 (Web Text))

Max Activating Token Index: 788

Text #10

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2815. Min Act: -0.0001

Data Index: 1408035 (Python Code)

Max Activating Token Index: 477

Text #11

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2831. Min Act: -0.0000

Data Index: 750930 (C4 (Web Text))

Max Activating Token Index: 156

Text #12

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2774. Min Act: -0.0000

Data Index: 737664 (C4 (Web Text))

Max Activating Token Index: 926

Text #13

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2667. Min Act: -0.0000

Data Index: 372590 (C4 (Web Text))

Max Activating Token Index: 755

Text #14

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2767. Min Act: -0.0000

Data Index: 239611 (C4 (Web Text))

Max Activating Token Index: 918

Text #15

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2582. Min Act: -0.0000

Data Index: 59828 (C4 (Web Text))

Max Activating Token Index: 555

Text #16

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2833. Min Act: -0.0000

Data Index: 54310 (C4 (Web Text))

Max Activating Token Index: 271

Text #17

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2659. Min Act: -0.0000

Data Index: 19911 (C4 (Web Text))

Max Activating Token Index: 273

Text #18

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2741. Min Act: -0.0001

Data Index: 1370169 (Python Code)

Max Activating Token Index: 990

Text #19

Max Range: 0.4505. Min Range: -0.4505

Max Act: 0.2639. Min Act: -0.0000

Data Index: 931056 (C4 (Web Text))

Max Activating Token Index: 995