Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1088 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 5.1265. Min Range: -5.1265

Max Act: 5.0679. Min Act: -0.0001

Data Index: 287053 (C4 (Web Text))

Max Activating Token Index: 48

Text #1

Max Range: 5.1265. Min Range: -5.1265

Max Act: 5.1265. Min Act: -0.0001

Data Index: 1647059 (Python Code)

Max Activating Token Index: 572

Text #2

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.8339. Min Act: -0.0001

Data Index: 402281 (C4 (Web Text))

Max Activating Token Index: 149

Text #3

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.6121. Min Act: -0.0001

Data Index: 1531829 (Python Code)

Max Activating Token Index: 449

Text #4

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.4610. Min Act: -0.0001

Data Index: 66754 (C4 (Web Text))

Max Activating Token Index: 637

Text #5

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.4338. Min Act: -0.0001

Data Index: 1332389 (C4 (Web Text))

Max Activating Token Index: 10

Text #6

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.9822. Min Act: -0.0001

Data Index: 1655923 (Python Code)

Max Activating Token Index: 482

Text #7

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.1492. Min Act: -0.0001

Data Index: 1573203 (Python Code)

Max Activating Token Index: 739

Text #8

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.9386. Min Act: -0.0001

Data Index: 1589620 (Python Code)

Max Activating Token Index: 868

Text #9

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.0274. Min Act: -0.0001

Data Index: 1604525 (Python Code)

Max Activating Token Index: 8

Text #10

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.7117. Min Act: -0.0001

Data Index: 1370160 (Python Code)

Max Activating Token Index: 430

Text #11

Max Range: 5.1265. Min Range: -5.1265

Max Act: 4.0253. Min Act: -0.0001

Data Index: 1620477 (Python Code)

Max Activating Token Index: 10

Text #12

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.6088. Min Act: -0.0001

Data Index: 1128102 (C4 (Web Text))

Max Activating Token Index: 345

Text #13

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.6762. Min Act: -0.0001

Data Index: 1528099 (Python Code)

Max Activating Token Index: 672

Text #14

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.5380. Min Act: -0.0001

Data Index: 1502789 (Python Code)

Max Activating Token Index: 635

Text #15

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.9974. Min Act: -0.0001

Data Index: 1098083 (C4 (Web Text))

Max Activating Token Index: 279

Text #16

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.8327. Min Act: -0.0001

Data Index: 1550839 (Python Code)

Max Activating Token Index: 353

Text #17

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.9830. Min Act: -0.0001

Data Index: 1386801 (Python Code)

Max Activating Token Index: 612

Text #18

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.5123. Min Act: -0.0001

Data Index: 1484361 (Python Code)

Max Activating Token Index: 326

Text #19

Max Range: 5.1265. Min Range: -5.1265

Max Act: 3.6415. Min Act: -0.0001

Data Index: 1526196 (Python Code)

Max Activating Token Index: 453