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

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

Neuron 1900 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1931. Min Range: -2.1931

Max Act: 2.0908. Min Act: -0.0001

Data Index: 1404429 (Python Code)

Max Activating Token Index: 1005

Text #1

Max Range: 2.1931. Min Range: -2.1931

Max Act: 2.1931. Min Act: -0.0001

Data Index: 878219 (C4 (Web Text))

Max Activating Token Index: 724

Text #2

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.9994. Min Act: -0.0001

Data Index: 1528118 (Python Code)

Max Activating Token Index: 973

Text #3

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.9176. Min Act: -0.0000

Data Index: 1393759 (Python Code)

Max Activating Token Index: 1004

Text #4

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.8143. Min Act: -0.0001

Data Index: 384535 (C4 (Web Text))

Max Activating Token Index: 1020

Text #5

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.7570. Min Act: -0.0001

Data Index: 847215 (C4 (Web Text))

Max Activating Token Index: 689

Text #6

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.5979. Min Act: -0.0000

Data Index: 1514113 (Python Code)

Max Activating Token Index: 975

Text #7

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.6157. Min Act: -0.0001

Data Index: 1482094 (Python Code)

Max Activating Token Index: 862

Text #8

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.5318. Min Act: -0.0001

Data Index: 1489805 (Python Code)

Max Activating Token Index: 933

Text #9

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.4089. Min Act: -0.0000

Data Index: 1564090 (Python Code)

Max Activating Token Index: 1002

Text #10

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.4944. Min Act: -0.0000

Data Index: 1419834 (Python Code)

Max Activating Token Index: 747

Text #11

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.3531. Min Act: -0.0001

Data Index: 1589428 (Python Code)

Max Activating Token Index: 655

Text #12

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.2965. Min Act: -0.0000

Data Index: 1474464 (Python Code)

Max Activating Token Index: 854

Text #13

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.2780. Min Act: -0.0000

Data Index: 1577364 (Python Code)

Max Activating Token Index: 902

Text #14

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.1880. Min Act: -0.0000

Data Index: 1526982 (Python Code)

Max Activating Token Index: 240

Text #15

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.2583. Min Act: -0.0000

Data Index: 1564067 (Python Code)

Max Activating Token Index: 557

Text #16

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.2213. Min Act: -0.0000

Data Index: 1647679 (Python Code)

Max Activating Token Index: 848

Text #17

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.3014. Min Act: -0.0000

Data Index: 1393231 (Python Code)

Max Activating Token Index: 170

Text #18

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.1318. Min Act: -0.0000

Data Index: 1437428 (Python Code)

Max Activating Token Index: 520

Text #19

Max Range: 2.1931. Min Range: -2.1931

Max Act: 1.1810. Min Act: -0.0000

Data Index: 1382426 (Python Code)

Max Activating Token Index: 211