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

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

Neuron 3967 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: 2.4125. Min Range: -2.4125

Max Act: 2.0313. Min Act: -0.0000

Data Index: 1200658 (C4 (Web Text))

Max Activating Token Index: 928

Text #1

Max Range: 2.4125. Min Range: -2.4125

Max Act: 2.4125. Min Act: -0.0000

Data Index: 611713 (C4 (Web Text))

Max Activating Token Index: 1002

Text #2

Max Range: 2.4125. Min Range: -2.4125

Max Act: 2.2008. Min Act: -0.0000

Data Index: 1211353 (C4 (Web Text))

Max Activating Token Index: 851

Text #3

Max Range: 2.4125. Min Range: -2.4125

Max Act: 2.0001. Min Act: -0.0000

Data Index: 966230 (C4 (Web Text))

Max Activating Token Index: 869

Text #4

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.8199. Min Act: -0.0000

Data Index: 703001 (C4 (Web Text))

Max Activating Token Index: 761

Text #5

Max Range: 2.4125. Min Range: -2.4125

Max Act: 2.0297. Min Act: -0.0000

Data Index: 35213 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9099. Min Act: -0.0000

Data Index: 604863 (C4 (Web Text))

Max Activating Token Index: 683

Text #7

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9100. Min Act: -0.0000

Data Index: 432957 (C4 (Web Text))

Max Activating Token Index: 77

Text #8

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9138. Min Act: -0.0000

Data Index: 1560792 (Python Code)

Max Activating Token Index: 361

Text #9

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9039. Min Act: -0.0000

Data Index: 240834 (C4 (Web Text))

Max Activating Token Index: 956

Text #10

Max Range: 2.4125. Min Range: -2.4125

Max Act: 2.0237. Min Act: -0.0000

Data Index: 758315 (C4 (Web Text))

Max Activating Token Index: 194

Text #11

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9059. Min Act: -0.0000

Data Index: 1263282 (C4 (Web Text))

Max Activating Token Index: 479

Text #12

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.8605. Min Act: -0.0000

Data Index: 478605 (C4 (Web Text))

Max Activating Token Index: 844

Text #13

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.8939. Min Act: -0.0000

Data Index: 620121 (C4 (Web Text))

Max Activating Token Index: 1007

Text #14

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.9999. Min Act: -0.0000

Data Index: 634365 (C4 (Web Text))

Max Activating Token Index: 943

Text #15

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.7804. Min Act: -0.0000

Data Index: 1611248 (Python Code)

Max Activating Token Index: 82

Text #16

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.6685. Min Act: -0.0000

Data Index: 1267692 (C4 (Web Text))

Max Activating Token Index: 595

Text #17

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.7755. Min Act: -0.0000

Data Index: 261406 (C4 (Web Text))

Max Activating Token Index: 523

Text #18

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.8575. Min Act: -0.0000

Data Index: 1023321 (C4 (Web Text))

Max Activating Token Index: 719

Text #19

Max Range: 2.4125. Min Range: -2.4125

Max Act: 1.5785. Min Act: -0.0000

Data Index: 1617873 (Python Code)

Max Activating Token Index: 122