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

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

Neuron 5700 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.9918. Min Act: -0.0000

Data Index: 29850 (C4 (Web Text))

Max Activating Token Index: 807

Text #1

Max Range: 2.0823. Min Range: -2.0823

Max Act: 2.0823. Min Act: -0.0000

Data Index: 1426575 (Python Code)

Max Activating Token Index: 11

Text #2

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.9085. Min Act: -0.0000

Data Index: 1401141 (Python Code)

Max Activating Token Index: 10

Text #3

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.9085. Min Act: -0.0000

Data Index: 1596469 (Python Code)

Max Activating Token Index: 10

Text #4

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.9571. Min Act: -0.0000

Data Index: 1090865 (C4 (Web Text))

Max Activating Token Index: 854

Text #5

Max Range: 2.0823. Min Range: -2.0823

Max Act: 2.0368. Min Act: -0.0000

Data Index: 107112 (C4 (Web Text))

Max Activating Token Index: 1002

Text #6

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.8524. Min Act: -0.0000

Data Index: 802757 (C4 (Web Text))

Max Activating Token Index: 502

Text #7

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.7649. Min Act: -0.0000

Data Index: 1442003 (Python Code)

Max Activating Token Index: 7

Text #8

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.7837. Min Act: -0.0000

Data Index: 91960 (C4 (Web Text))

Max Activating Token Index: 1017

Text #9

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.7625. Min Act: -0.0000

Data Index: 812379 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.6887. Min Act: -0.0000

Data Index: 398386 (C4 (Web Text))

Max Activating Token Index: 201

Text #11

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.6547. Min Act: -0.0000

Data Index: 139965 (C4 (Web Text))

Max Activating Token Index: 683

Text #12

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.5391. Min Act: -0.0000

Data Index: 278436 (C4 (Web Text))

Max Activating Token Index: 67

Text #13

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.6180. Min Act: -0.0000

Data Index: 1435785 (Python Code)

Max Activating Token Index: 6

Text #14

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.5493. Min Act: -0.0000

Data Index: 1378894 (Python Code)

Max Activating Token Index: 17

Text #15

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.3977. Min Act: -0.0000

Data Index: 556770 (C4 (Web Text))

Max Activating Token Index: 308

Text #16

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.4119. Min Act: -0.0000

Data Index: 1089712 (C4 (Web Text))

Max Activating Token Index: 542

Text #17

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.6309. Min Act: -0.0000

Data Index: 1249462 (C4 (Web Text))

Max Activating Token Index: 830

Text #18

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.5355. Min Act: -0.0000

Data Index: 1204709 (C4 (Web Text))

Max Activating Token Index: 197

Text #19

Max Range: 2.0823. Min Range: -2.0823

Max Act: 1.5550. Min Act: -0.0000

Data Index: 305615 (C4 (Web Text))

Max Activating Token Index: 914