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

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

Neuron 3396 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.7516. Min Act: -0.0001

Data Index: 1641807 (Python Code)

Max Activating Token Index: 120

Text #1

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.8131. Min Act: -0.0001

Data Index: 1530624 (Python Code)

Max Activating Token Index: 666

Text #2

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.9445. Min Act: -0.0001

Data Index: 1504853 (Python Code)

Max Activating Token Index: 479

Text #3

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.8379. Min Act: -0.0001

Data Index: 1109952 (C4 (Web Text))

Max Activating Token Index: 206

Text #4

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.7676. Min Act: -0.0001

Data Index: 1656118 (Python Code)

Max Activating Token Index: 22

Text #5

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.7327. Min Act: -0.0001

Data Index: 1401110 (Python Code)

Max Activating Token Index: 176

Text #6

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6771. Min Act: -0.0001

Data Index: 1495232 (Python Code)

Max Activating Token Index: 105

Text #7

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6468. Min Act: -0.0001

Data Index: 1577994 (Python Code)

Max Activating Token Index: 101

Text #8

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6547. Min Act: -0.0001

Data Index: 1479798 (Python Code)

Max Activating Token Index: 687

Text #9

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6623. Min Act: -0.0001

Data Index: 1413612 (Python Code)

Max Activating Token Index: 969

Text #10

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6397. Min Act: -0.0001

Data Index: 1169340 (C4 (Web Text))

Max Activating Token Index: 207

Text #11

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.5608. Min Act: -0.0001

Data Index: 1400988 (Python Code)

Max Activating Token Index: 206

Text #12

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6150. Min Act: -0.0001

Data Index: 1533183 (Python Code)

Max Activating Token Index: 13

Text #13

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.5990. Min Act: -0.0001

Data Index: 1408427 (Python Code)

Max Activating Token Index: 849

Text #14

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6016. Min Act: -0.0001

Data Index: 1249386 (C4 (Web Text))

Max Activating Token Index: 645

Text #15

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6100. Min Act: -0.0001

Data Index: 1529782 (Python Code)

Max Activating Token Index: 337

Text #16

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.6111. Min Act: -0.0001

Data Index: 1563720 (Python Code)

Max Activating Token Index: 342

Text #17

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.5772. Min Act: -0.0001

Data Index: 632219 (C4 (Web Text))

Max Activating Token Index: 164

Text #18

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.5751. Min Act: -0.0001

Data Index: 1567690 (Python Code)

Max Activating Token Index: 15

Text #19

Max Range: 0.9445. Min Range: -0.9445

Max Act: 0.5730. Min Act: -0.0001

Data Index: 525043 (C4 (Web Text))

Max Activating Token Index: 783