Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3637 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.9825. Min Act: -0.0001

Data Index: 35008 (C4 (Web Text))

Max Activating Token Index: 303

Text #1

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8514. Min Act: -0.0001

Data Index: 1315755 (C4 (Web Text))

Max Activating Token Index: 222

Text #2

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.9141. Min Act: -0.0001

Data Index: 866079 (C4 (Web Text))

Max Activating Token Index: 915

Text #3

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8166. Min Act: -0.0001

Data Index: 1234206 (C4 (Web Text))

Max Activating Token Index: 460

Text #4

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8835. Min Act: -0.0001

Data Index: 544723 (C4 (Web Text))

Max Activating Token Index: 238

Text #5

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8242. Min Act: -0.0001

Data Index: 41585 (C4 (Web Text))

Max Activating Token Index: 989

Text #6

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8833. Min Act: -0.0001

Data Index: 1635645 (Python Code)

Max Activating Token Index: 96

Text #7

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7969. Min Act: -0.0001

Data Index: 43811 (C4 (Web Text))

Max Activating Token Index: 218

Text #8

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8395. Min Act: -0.0001

Data Index: 551612 (C4 (Web Text))

Max Activating Token Index: 969

Text #9

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7588. Min Act: -0.0001

Data Index: 1193384 (C4 (Web Text))

Max Activating Token Index: 450

Text #10

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7908. Min Act: -0.0001

Data Index: 596609 (C4 (Web Text))

Max Activating Token Index: 485

Text #11

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7267. Min Act: -0.0001

Data Index: 918473 (C4 (Web Text))

Max Activating Token Index: 271

Text #12

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7829. Min Act: -0.0001

Data Index: 37989 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.8322. Min Act: -0.0001

Data Index: 100373 (C4 (Web Text))

Max Activating Token Index: 117

Text #14

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7888. Min Act: -0.0001

Data Index: 75176 (C4 (Web Text))

Max Activating Token Index: 616

Text #15

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7499. Min Act: -0.0001

Data Index: 1275541 (C4 (Web Text))

Max Activating Token Index: 9

Text #16

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7532. Min Act: -0.0001

Data Index: 937681 (C4 (Web Text))

Max Activating Token Index: 542

Text #17

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7234. Min Act: -0.0001

Data Index: 244600 (C4 (Web Text))

Max Activating Token Index: 394

Text #18

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7734. Min Act: -0.0001

Data Index: 133436 (C4 (Web Text))

Max Activating Token Index: 521

Text #19

Max Range: 0.9825. Min Range: -0.9825

Max Act: 0.7070. Min Act: -0.0001

Data Index: 53359 (C4 (Web Text))

Max Activating Token Index: 930