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

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

Neuron 1610 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.6245. Min Act: -0.0000

Data Index: 308344 (C4 (Web Text))

Max Activating Token Index: 787

Text #1

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.7919. Min Act: -0.0000

Data Index: 1289842 (C4 (Web Text))

Max Activating Token Index: 179

Text #2

Max Range: 4.0840. Min Range: -4.0840

Max Act: 4.0840. Min Act: -0.0000

Data Index: 886688 (C4 (Web Text))

Max Activating Token Index: 904

Text #3

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.1642. Min Act: -0.0000

Data Index: 440308 (C4 (Web Text))

Max Activating Token Index: 738

Text #4

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.4597. Min Act: -0.0000

Data Index: 1060960 (C4 (Web Text))

Max Activating Token Index: 723

Text #5

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.5723. Min Act: -0.0000

Data Index: 1049479 (C4 (Web Text))

Max Activating Token Index: 301

Text #6

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.1182. Min Act: -0.0000

Data Index: 1401740 (Python Code)

Max Activating Token Index: 1008

Text #7

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.9590. Min Act: -0.0000

Data Index: 1572350 (Python Code)

Max Activating Token Index: 163

Text #8

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.1784. Min Act: -0.0000

Data Index: 399181 (C4 (Web Text))

Max Activating Token Index: 874

Text #9

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.8961. Min Act: -0.0000

Data Index: 167744 (C4 (Web Text))

Max Activating Token Index: 546

Text #10

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.9774. Min Act: -0.0000

Data Index: 875409 (C4 (Web Text))

Max Activating Token Index: 910

Text #11

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.0019. Min Act: -0.0000

Data Index: 267685 (C4 (Web Text))

Max Activating Token Index: 338

Text #12

Max Range: 4.0840. Min Range: -4.0840

Max Act: 3.1535. Min Act: -0.0000

Data Index: 1555537 (Python Code)

Max Activating Token Index: 190

Text #13

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.8240. Min Act: -0.0000

Data Index: 1386009 (Python Code)

Max Activating Token Index: 447

Text #14

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.9441. Min Act: -0.0000

Data Index: 978733 (C4 (Web Text))

Max Activating Token Index: 679

Text #15

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.9312. Min Act: -0.0000

Data Index: 1316156 (C4 (Web Text))

Max Activating Token Index: 403

Text #16

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.7762. Min Act: -0.0000

Data Index: 948191 (C4 (Web Text))

Max Activating Token Index: 126

Text #17

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.7617. Min Act: -0.0000

Data Index: 380350 (C4 (Web Text))

Max Activating Token Index: 838

Text #18

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.8690. Min Act: -0.0000

Data Index: 797131 (C4 (Web Text))

Max Activating Token Index: 836

Text #19

Max Range: 4.0840. Min Range: -4.0840

Max Act: 2.8442. Min Act: -0.0000

Data Index: 687495 (C4 (Web Text))

Max Activating Token Index: 667