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

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

Neuron 3829 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.9145. Min Act: -0.0001

Data Index: 477258 (C4 (Web Text))

Max Activating Token Index: 177

Text #1

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8735. Min Act: -0.0001

Data Index: 1458455 (Python Code)

Max Activating Token Index: 174

Text #2

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8283. Min Act: -0.0001

Data Index: 1529757 (Python Code)

Max Activating Token Index: 3

Text #3

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8305. Min Act: -0.0001

Data Index: 746350 (C4 (Web Text))

Max Activating Token Index: 248

Text #4

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7977. Min Act: -0.0001

Data Index: 1645930 (Python Code)

Max Activating Token Index: 447

Text #5

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8679. Min Act: -0.0001

Data Index: 1435359 (Python Code)

Max Activating Token Index: 151

Text #6

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8581. Min Act: -0.0001

Data Index: 684464 (C4 (Web Text))

Max Activating Token Index: 271

Text #7

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8521. Min Act: -0.0001

Data Index: 1649868 (Python Code)

Max Activating Token Index: 365

Text #8

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7691. Min Act: -0.0001

Data Index: 1464758 (Python Code)

Max Activating Token Index: 781

Text #9

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7901. Min Act: -0.0001

Data Index: 471845 (C4 (Web Text))

Max Activating Token Index: 152

Text #10

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8511. Min Act: -0.0001

Data Index: 1438392 (Python Code)

Max Activating Token Index: 374

Text #11

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7550. Min Act: -0.0001

Data Index: 1215148 (C4 (Web Text))

Max Activating Token Index: 724

Text #12

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7502. Min Act: -0.0001

Data Index: 1413314 (Python Code)

Max Activating Token Index: 76

Text #13

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7901. Min Act: -0.0001

Data Index: 1617361 (Python Code)

Max Activating Token Index: 77

Text #14

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8239. Min Act: -0.0001

Data Index: 1429024 (Python Code)

Max Activating Token Index: 519

Text #15

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7640. Min Act: -0.0001

Data Index: 1374448 (Python Code)

Max Activating Token Index: 136

Text #16

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7627. Min Act: -0.0001

Data Index: 1128507 (C4 (Web Text))

Max Activating Token Index: 286

Text #17

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7764. Min Act: -0.0001

Data Index: 841733 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.7706. Min Act: -0.0001

Data Index: 1638902 (Python Code)

Max Activating Token Index: 889

Text #19

Max Range: 0.9145. Min Range: -0.9145

Max Act: 0.8677. Min Act: -0.0001

Data Index: 1575895 (Python Code)

Max Activating Token Index: 197