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

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

Neuron 4722 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.8124. Min Act: -0.0001

Data Index: 1601182 (Python Code)

Max Activating Token Index: 73

Text #1

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.7760. Min Act: -0.0001

Data Index: 1454638 (Python Code)

Max Activating Token Index: 133

Text #2

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6238. Min Act: -0.0001

Data Index: 1399729 (Python Code)

Max Activating Token Index: 711

Text #3

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6637. Min Act: -0.0001

Data Index: 1594335 (Python Code)

Max Activating Token Index: 29

Text #4

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.7223. Min Act: -0.0001

Data Index: 1414819 (Python Code)

Max Activating Token Index: 240

Text #5

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6176. Min Act: -0.0001

Data Index: 1384247 (Python Code)

Max Activating Token Index: 235

Text #6

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6361. Min Act: -0.0001

Data Index: 599181 (C4 (Web Text))

Max Activating Token Index: 45

Text #7

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6501. Min Act: -0.0001

Data Index: 1603054 (Python Code)

Max Activating Token Index: 461

Text #8

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6059. Min Act: -0.0001

Data Index: 1616639 (Python Code)

Max Activating Token Index: 80

Text #9

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6028. Min Act: -0.0001

Data Index: 1590826 (Python Code)

Max Activating Token Index: 882

Text #10

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5693. Min Act: -0.0001

Data Index: 1431285 (Python Code)

Max Activating Token Index: 434

Text #11

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5746. Min Act: -0.0001

Data Index: 1462339 (Python Code)

Max Activating Token Index: 337

Text #12

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6027. Min Act: -0.0001

Data Index: 1493353 (Python Code)

Max Activating Token Index: 83

Text #13

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5682. Min Act: -0.0001

Data Index: 1385876 (Python Code)

Max Activating Token Index: 114

Text #14

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5639. Min Act: -0.0001

Data Index: 1550076 (Python Code)

Max Activating Token Index: 943

Text #15

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5748. Min Act: -0.0001

Data Index: 1653597 (Python Code)

Max Activating Token Index: 189

Text #16

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5879. Min Act: -0.0001

Data Index: 70065 (C4 (Web Text))

Max Activating Token Index: 23

Text #17

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.6058. Min Act: -0.0001

Data Index: 1619824 (Python Code)

Max Activating Token Index: 795

Text #18

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5748. Min Act: -0.0001

Data Index: 1619503 (Python Code)

Max Activating Token Index: 433

Text #19

Max Range: 0.8124. Min Range: -0.8124

Max Act: 0.5622. Min Act: -0.0001

Data Index: 1408957 (Python Code)

Max Activating Token Index: 560