Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3748 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.4668. Min Act: -0.0001

Data Index: 1416240 (Python Code)

Max Activating Token Index: 187

Text #1

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.5840. Min Act: -0.0000

Data Index: 20010 (C4 (Web Text))

Max Activating Token Index: 468

Text #2

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.2923. Min Act: -0.0000

Data Index: 1314989 (C4 (Web Text))

Max Activating Token Index: 433

Text #3

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.2707. Min Act: -0.0001

Data Index: 1009428 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.1674. Min Act: -0.0001

Data Index: 1432531 (Python Code)

Max Activating Token Index: 631

Text #5

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.1772. Min Act: -0.0001

Data Index: 1387497 (Python Code)

Max Activating Token Index: 761

Text #6

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.2232. Min Act: -0.0001

Data Index: 1383046 (Python Code)

Max Activating Token Index: 225

Text #7

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0015. Min Act: -0.0001

Data Index: 1398500 (Python Code)

Max Activating Token Index: 655

Text #8

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.1011. Min Act: -0.0001

Data Index: 1418805 (Python Code)

Max Activating Token Index: 389

Text #9

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0247. Min Act: -0.0001

Data Index: 1195522 (C4 (Web Text))

Max Activating Token Index: 188

Text #10

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0301. Min Act: -0.0001

Data Index: 1639414 (Python Code)

Max Activating Token Index: 639

Text #11

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.2236. Min Act: -0.0001

Data Index: 1626488 (Python Code)

Max Activating Token Index: 537

Text #12

Max Range: 1.5840. Min Range: -1.5840

Max Act: 0.9795. Min Act: -0.0001

Data Index: 1570660 (Python Code)

Max Activating Token Index: 832

Text #13

Max Range: 1.5840. Min Range: -1.5840

Max Act: 0.9428. Min Act: -0.0001

Data Index: 1600678 (Python Code)

Max Activating Token Index: 101

Text #14

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0955. Min Act: -0.0001

Data Index: 1440957 (Python Code)

Max Activating Token Index: 635

Text #15

Max Range: 1.5840. Min Range: -1.5840

Max Act: 0.9830. Min Act: -0.0001

Data Index: 1522131 (Python Code)

Max Activating Token Index: 700

Text #16

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0436. Min Act: -0.0001

Data Index: 1409649 (Python Code)

Max Activating Token Index: 124

Text #17

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0174. Min Act: -0.0001

Data Index: 1362145 (Python Code)

Max Activating Token Index: 464

Text #18

Max Range: 1.5840. Min Range: -1.5840

Max Act: 1.0047. Min Act: -0.0001

Data Index: 1584375 (Python Code)

Max Activating Token Index: 802

Text #19

Max Range: 1.5840. Min Range: -1.5840

Max Act: 0.9981. Min Act: -0.0001

Data Index: 1367161 (Python Code)

Max Activating Token Index: 331