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

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

Neuron 3492 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.8024. Min Act: -0.0003

Data Index: 873394 (C4 (Web Text))

Max Activating Token Index: 902

Text #1

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7350. Min Act: -0.0003

Data Index: 1217479 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6866. Min Act: -0.0003

Data Index: 42192 (C4 (Web Text))

Max Activating Token Index: 627

Text #3

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7521. Min Act: -0.0003

Data Index: 50551 (C4 (Web Text))

Max Activating Token Index: 821

Text #4

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7486. Min Act: -0.0003

Data Index: 1042747 (C4 (Web Text))

Max Activating Token Index: 40

Text #5

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7222. Min Act: -0.0003

Data Index: 81889 (C4 (Web Text))

Max Activating Token Index: 28

Text #6

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7267. Min Act: -0.0003

Data Index: 382131 (C4 (Web Text))

Max Activating Token Index: 965

Text #7

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6851. Min Act: -0.0003

Data Index: 1249709 (C4 (Web Text))

Max Activating Token Index: 114

Text #8

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6850. Min Act: -0.0003

Data Index: 1047027 (C4 (Web Text))

Max Activating Token Index: 114

Text #9

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7085. Min Act: -0.0003

Data Index: 922476 (C4 (Web Text))

Max Activating Token Index: 49

Text #10

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7083. Min Act: -0.0003

Data Index: 100555 (C4 (Web Text))

Max Activating Token Index: 771

Text #11

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6636. Min Act: -0.0003

Data Index: 378985 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7116. Min Act: -0.0003

Data Index: 238655 (C4 (Web Text))

Max Activating Token Index: 799

Text #13

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7063. Min Act: -0.0003

Data Index: 466567 (C4 (Web Text))

Max Activating Token Index: 97

Text #14

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6798. Min Act: -0.0003

Data Index: 625742 (C4 (Web Text))

Max Activating Token Index: 91

Text #15

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6983. Min Act: -0.0003

Data Index: 634415 (C4 (Web Text))

Max Activating Token Index: 470

Text #16

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6523. Min Act: -0.0003

Data Index: 456351 (C4 (Web Text))

Max Activating Token Index: 555

Text #17

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6699. Min Act: -0.0003

Data Index: 200307 (C4 (Web Text))

Max Activating Token Index: 730

Text #18

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.6725. Min Act: -0.0003

Data Index: 1151553 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 0.8024. Min Range: -0.8024

Max Act: 0.7402. Min Act: -0.0003

Data Index: 489692 (C4 (Web Text))

Max Activating Token Index: 273