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

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

Neuron 5033 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.7788. Min Range: -4.7788

Max Act: 4.7788. Min Act: -0.0000

Data Index: 1431382 (Python Code)

Max Activating Token Index: 150

Text #1

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.7834. Min Act: -0.0000

Data Index: 1416295 (Python Code)

Max Activating Token Index: 38

Text #2

Max Range: 4.7788. Min Range: -4.7788

Max Act: 4.0136. Min Act: -0.0000

Data Index: 627337 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.6577. Min Act: -0.0000

Data Index: 682242 (C4 (Web Text))

Max Activating Token Index: 48

Text #4

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.4448. Min Act: -0.0000

Data Index: 1638708 (Python Code)

Max Activating Token Index: 32

Text #5

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.3934. Min Act: -0.0000

Data Index: 1276950 (C4 (Web Text))

Max Activating Token Index: 9

Text #6

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.3764. Min Act: -0.0000

Data Index: 1203296 (C4 (Web Text))

Max Activating Token Index: 208

Text #7

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.2768. Min Act: -0.0000

Data Index: 677066 (C4 (Web Text))

Max Activating Token Index: 48

Text #8

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.2907. Min Act: -0.0000

Data Index: 1240850 (C4 (Web Text))

Max Activating Token Index: 42

Text #9

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.4462. Min Act: -0.0000

Data Index: 303238 (C4 (Web Text))

Max Activating Token Index: 35

Text #10

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.2710. Min Act: -0.0000

Data Index: 178863 (C4 (Web Text))

Max Activating Token Index: 508

Text #11

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.3931. Min Act: -0.0000

Data Index: 1480370 (Python Code)

Max Activating Token Index: 356

Text #12

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.1911. Min Act: -0.0000

Data Index: 1607711 (Python Code)

Max Activating Token Index: 295

Text #13

Max Range: 4.7788. Min Range: -4.7788

Max Act: 2.9816. Min Act: -0.0000

Data Index: 208214 (C4 (Web Text))

Max Activating Token Index: 577

Text #14

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.1374. Min Act: -0.0000

Data Index: 1499598 (Python Code)

Max Activating Token Index: 1011

Text #15

Max Range: 4.7788. Min Range: -4.7788

Max Act: 2.9429. Min Act: -0.0000

Data Index: 1305081 (C4 (Web Text))

Max Activating Token Index: 52

Text #16

Max Range: 4.7788. Min Range: -4.7788

Max Act: 2.9539. Min Act: -0.0000

Data Index: 704345 (C4 (Web Text))

Max Activating Token Index: 198

Text #17

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.2434. Min Act: -0.0000

Data Index: 1265399 (C4 (Web Text))

Max Activating Token Index: 81

Text #18

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.0326. Min Act: -0.0000

Data Index: 47664 (C4 (Web Text))

Max Activating Token Index: 459

Text #19

Max Range: 4.7788. Min Range: -4.7788

Max Act: 3.0419. Min Act: -0.0000

Data Index: 1323912 (C4 (Web Text))

Max Activating Token Index: 835