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

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

Neuron 3379 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.7755. Min Act: -0.0000

Data Index: 412333 (C4 (Web Text))

Max Activating Token Index: 350

Text #1

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5532. Min Act: -0.0001

Data Index: 151290 (C4 (Web Text))

Max Activating Token Index: 616

Text #2

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5434. Min Act: -0.0001

Data Index: 1267081 (C4 (Web Text))

Max Activating Token Index: 292

Text #3

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5153. Min Act: -0.0001

Data Index: 1395600 (Python Code)

Max Activating Token Index: 764

Text #4

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5159. Min Act: -0.0001

Data Index: 863618 (C4 (Web Text))

Max Activating Token Index: 734

Text #5

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4813. Min Act: -0.0001

Data Index: 1516073 (Python Code)

Max Activating Token Index: 481

Text #6

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5152. Min Act: -0.0001

Data Index: 653938 (C4 (Web Text))

Max Activating Token Index: 731

Text #7

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4460. Min Act: -0.0001

Data Index: 1318632 (C4 (Web Text))

Max Activating Token Index: 824

Text #8

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4400. Min Act: -0.0001

Data Index: 1577134 (Python Code)

Max Activating Token Index: 754

Text #9

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4538. Min Act: -0.0001

Data Index: 256252 (C4 (Web Text))

Max Activating Token Index: 498

Text #10

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4285. Min Act: -0.0001

Data Index: 462979 (C4 (Web Text))

Max Activating Token Index: 1003

Text #11

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.5051. Min Act: -0.0001

Data Index: 1259649 (C4 (Web Text))

Max Activating Token Index: 767

Text #12

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4377. Min Act: -0.0001

Data Index: 1036897 (C4 (Web Text))

Max Activating Token Index: 757

Text #13

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4198. Min Act: -0.0000

Data Index: 464553 (C4 (Web Text))

Max Activating Token Index: 78

Text #14

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4100. Min Act: -0.0001

Data Index: 329450 (C4 (Web Text))

Max Activating Token Index: 103

Text #15

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4091. Min Act: -0.0000

Data Index: 1400582 (Python Code)

Max Activating Token Index: 728

Text #16

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.3967. Min Act: -0.0001

Data Index: 91722 (C4 (Web Text))

Max Activating Token Index: 920

Text #17

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4040. Min Act: -0.0001

Data Index: 113802 (C4 (Web Text))

Max Activating Token Index: 562

Text #18

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4005. Min Act: -0.0001

Data Index: 1057135 (C4 (Web Text))

Max Activating Token Index: 77

Text #19

Max Range: 0.7755. Min Range: -0.7755

Max Act: 0.4112. Min Act: -0.0000

Data Index: 1770 (C4 (Web Text))

Max Activating Token Index: 112