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

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

Neuron 4907 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.5665. Min Act: -0.0000

Data Index: 68124 (C4 (Web Text))

Max Activating Token Index: 186

Text #1

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.3450. Min Act: -0.0000

Data Index: 1126865 (C4 (Web Text))

Max Activating Token Index: 67

Text #2

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.3201. Min Act: -0.0000

Data Index: 1063073 (C4 (Web Text))

Max Activating Token Index: 329

Text #3

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.3175. Min Act: -0.0000

Data Index: 192663 (C4 (Web Text))

Max Activating Token Index: 994

Text #4

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2956. Min Act: -0.0000

Data Index: 489668 (C4 (Web Text))

Max Activating Token Index: 736

Text #5

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2861. Min Act: -0.0000

Data Index: 711306 (C4 (Web Text))

Max Activating Token Index: 683

Text #6

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2913. Min Act: -0.0000

Data Index: 903231 (C4 (Web Text))

Max Activating Token Index: 404

Text #7

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.3012. Min Act: -0.0000

Data Index: 760627 (C4 (Web Text))

Max Activating Token Index: 864

Text #8

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2777. Min Act: -0.0000

Data Index: 22384 (C4 (Web Text))

Max Activating Token Index: 962

Text #9

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2772. Min Act: -0.0000

Data Index: 310599 (C4 (Web Text))

Max Activating Token Index: 60

Text #10

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2745. Min Act: -0.0000

Data Index: 701755 (C4 (Web Text))

Max Activating Token Index: 33

Text #11

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2687. Min Act: -0.0000

Data Index: 1415834 (Python Code)

Max Activating Token Index: 202

Text #12

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2796. Min Act: -0.0000

Data Index: 1057577 (C4 (Web Text))

Max Activating Token Index: 956

Text #13

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2614. Min Act: -0.0000

Data Index: 986834 (C4 (Web Text))

Max Activating Token Index: 135

Text #14

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2479. Min Act: -0.0000

Data Index: 842986 (C4 (Web Text))

Max Activating Token Index: 988

Text #15

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2542. Min Act: -0.0000

Data Index: 359240 (C4 (Web Text))

Max Activating Token Index: 854

Text #16

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2524. Min Act: -0.0000

Data Index: 1352551 (C4 (Web Text))

Max Activating Token Index: 609

Text #17

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2603. Min Act: -0.0000

Data Index: 862873 (C4 (Web Text))

Max Activating Token Index: 867

Text #18

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2523. Min Act: -0.0000

Data Index: 1333937 (C4 (Web Text))

Max Activating Token Index: 45

Text #19

Max Range: 0.5665. Min Range: -0.5665

Max Act: 0.2409. Min Act: -0.0000

Data Index: 1049329 (C4 (Web Text))

Max Activating Token Index: 681