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

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

Neuron 3629 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.3817. Min Act: -0.0000

Data Index: 1634207 (Python Code)

Max Activating Token Index: 914

Text #1

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.3843. Min Act: -0.0000

Data Index: 7055 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.3775. Min Act: -0.0000

Data Index: 454964 (C4 (Web Text))

Max Activating Token Index: 907

Text #3

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.2638. Min Act: -0.0000

Data Index: 1465762 (Python Code)

Max Activating Token Index: 943

Text #4

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1471. Min Act: -0.0000

Data Index: 1519719 (Python Code)

Max Activating Token Index: 889

Text #5

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.2150. Min Act: -0.0000

Data Index: 452457 (C4 (Web Text))

Max Activating Token Index: 453

Text #6

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.2055. Min Act: -0.0000

Data Index: 959481 (C4 (Web Text))

Max Activating Token Index: 826

Text #7

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1179. Min Act: -0.0000

Data Index: 49733 (C4 (Web Text))

Max Activating Token Index: 334

Text #8

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.2569. Min Act: -0.0000

Data Index: 1160468 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1818. Min Act: -0.0000

Data Index: 931940 (C4 (Web Text))

Max Activating Token Index: 218

Text #10

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1759. Min Act: -0.0000

Data Index: 860615 (C4 (Web Text))

Max Activating Token Index: 306

Text #11

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1583. Min Act: -0.0000

Data Index: 1220370 (C4 (Web Text))

Max Activating Token Index: 839

Text #12

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.2236. Min Act: -0.0000

Data Index: 773742 (C4 (Web Text))

Max Activating Token Index: 964

Text #13

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1341. Min Act: -0.0000

Data Index: 412575 (C4 (Web Text))

Max Activating Token Index: 469

Text #14

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.0276. Min Act: -0.0000

Data Index: 1071614 (C4 (Web Text))

Max Activating Token Index: 863

Text #15

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1903. Min Act: -0.0000

Data Index: 1528098 (Python Code)

Max Activating Token Index: 975

Text #16

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1771. Min Act: -0.0000

Data Index: 53326 (C4 (Web Text))

Max Activating Token Index: 1014

Text #17

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1009. Min Act: -0.0000

Data Index: 1327092 (C4 (Web Text))

Max Activating Token Index: 264

Text #18

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.0709. Min Act: -0.0000

Data Index: 844857 (C4 (Web Text))

Max Activating Token Index: 61

Text #19

Max Range: 1.3843. Min Range: -1.3843

Max Act: 1.1250. Min Act: -0.0000

Data Index: 1248395 (C4 (Web Text))

Max Activating Token Index: 681