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

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

Neuron 631 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.5930. Min Act: -0.0001

Data Index: 469207 (C4 (Web Text))

Max Activating Token Index: 778

Text #1

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4914. Min Act: -0.0001

Data Index: 693848 (C4 (Web Text))

Max Activating Token Index: 670

Text #2

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4925. Min Act: -0.0001

Data Index: 344501 (C4 (Web Text))

Max Activating Token Index: 558

Text #3

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4408. Min Act: -0.0001

Data Index: 1307222 (C4 (Web Text))

Max Activating Token Index: 601

Text #4

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.5076. Min Act: -0.0001

Data Index: 419112 (C4 (Web Text))

Max Activating Token Index: 891

Text #5

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4699. Min Act: -0.0001

Data Index: 577643 (C4 (Web Text))

Max Activating Token Index: 848

Text #6

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4446. Min Act: -0.0001

Data Index: 378506 (C4 (Web Text))

Max Activating Token Index: 911

Text #7

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.5067. Min Act: -0.0001

Data Index: 1627370 (Python Code)

Max Activating Token Index: 737

Text #8

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4627. Min Act: -0.0001

Data Index: 1371890 (Python Code)

Max Activating Token Index: 48

Text #9

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4722. Min Act: -0.0001

Data Index: 1632582 (Python Code)

Max Activating Token Index: 139

Text #10

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4574. Min Act: -0.0001

Data Index: 642423 (C4 (Web Text))

Max Activating Token Index: 771

Text #11

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4289. Min Act: -0.0001

Data Index: 1271359 (C4 (Web Text))

Max Activating Token Index: 858

Text #12

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4606. Min Act: -0.0001

Data Index: 1051760 (C4 (Web Text))

Max Activating Token Index: 700

Text #13

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4318. Min Act: -0.0001

Data Index: 759951 (C4 (Web Text))

Max Activating Token Index: 527

Text #14

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4682. Min Act: -0.0001

Data Index: 720467 (C4 (Web Text))

Max Activating Token Index: 821

Text #15

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4359. Min Act: -0.0001

Data Index: 1173356 (C4 (Web Text))

Max Activating Token Index: 990

Text #16

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4520. Min Act: -0.0001

Data Index: 347064 (C4 (Web Text))

Max Activating Token Index: 351

Text #17

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4536. Min Act: -0.0001

Data Index: 600669 (C4 (Web Text))

Max Activating Token Index: 903

Text #18

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4345. Min Act: -0.0001

Data Index: 664427 (C4 (Web Text))

Max Activating Token Index: 908

Text #19

Max Range: 0.5930. Min Range: -0.5930

Max Act: 0.4483. Min Act: -0.0001

Data Index: 135144 (C4 (Web Text))

Max Activating Token Index: 85