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

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

Neuron 930 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: 0.7815. Min Range: -0.7815

Max Act: 0.7815. Min Act: -0.0001

Data Index: 1599641 (Python Code)

Max Activating Token Index: 778

Text #1

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.6565. Min Act: -0.0001

Data Index: 1468143 (Python Code)

Max Activating Token Index: 504

Text #2

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.6449. Min Act: -0.0001

Data Index: 238393 (C4 (Web Text))

Max Activating Token Index: 221

Text #3

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.6294. Min Act: -0.0001

Data Index: 1410419 (Python Code)

Max Activating Token Index: 791

Text #4

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5884. Min Act: -0.0001

Data Index: 1461261 (Python Code)

Max Activating Token Index: 535

Text #5

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5657. Min Act: -0.0001

Data Index: 155221 (C4 (Web Text))

Max Activating Token Index: 644

Text #6

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5651. Min Act: -0.0000

Data Index: 1429446 (Python Code)

Max Activating Token Index: 725

Text #7

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5923. Min Act: -0.0001

Data Index: 485420 (C4 (Web Text))

Max Activating Token Index: 197

Text #8

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5665. Min Act: -0.0001

Data Index: 1637796 (Python Code)

Max Activating Token Index: 471

Text #9

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5848. Min Act: -0.0001

Data Index: 1109561 (C4 (Web Text))

Max Activating Token Index: 917

Text #10

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5870. Min Act: -0.0000

Data Index: 707361 (C4 (Web Text))

Max Activating Token Index: 363

Text #11

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5567. Min Act: -0.0001

Data Index: 1349757 (C4 (Web Text))

Max Activating Token Index: 679

Text #12

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5747. Min Act: -0.0001

Data Index: 1540815 (Python Code)

Max Activating Token Index: 539

Text #13

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5456. Min Act: -0.0001

Data Index: 587444 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5153. Min Act: -0.0001

Data Index: 717277 (C4 (Web Text))

Max Activating Token Index: 349

Text #15

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5474. Min Act: -0.0001

Data Index: 1459599 (Python Code)

Max Activating Token Index: 523

Text #16

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5231. Min Act: -0.0000

Data Index: 1004379 (C4 (Web Text))

Max Activating Token Index: 606

Text #17

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5368. Min Act: -0.0000

Data Index: 747968 (C4 (Web Text))

Max Activating Token Index: 637

Text #18

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5208. Min Act: -0.0000

Data Index: 784784 (C4 (Web Text))

Max Activating Token Index: 241

Text #19

Max Range: 0.7815. Min Range: -0.7815

Max Act: 0.5306. Min Act: -0.0001

Data Index: 685604 (C4 (Web Text))

Max Activating Token Index: 99