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

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

Neuron 973 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: 0.0705. Min Range: -0.0705

Max Act: 0.0705. Min Act: -0.0001

Data Index: 1620162 (Python Code)

Max Activating Token Index: 897

Text #1

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0702. Min Act: -0.0001

Data Index: 614397 (C4 (Web Text))

Max Activating Token Index: 620

Text #2

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0667. Min Act: -0.0001

Data Index: 112144 (C4 (Web Text))

Max Activating Token Index: 241

Text #3

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0659. Min Act: -0.0001

Data Index: 1430522 (Python Code)

Max Activating Token Index: 532

Text #4

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0674. Min Act: -0.0001

Data Index: 357970 (C4 (Web Text))

Max Activating Token Index: 432

Text #5

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0597. Min Act: -0.0001

Data Index: 1363438 (Python Code)

Max Activating Token Index: 242

Text #6

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0590. Min Act: -0.0001

Data Index: 1653612 (Python Code)

Max Activating Token Index: 465

Text #7

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0599. Min Act: -0.0001

Data Index: 871094 (C4 (Web Text))

Max Activating Token Index: 640

Text #8

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0576. Min Act: -0.0001

Data Index: 1158689 (C4 (Web Text))

Max Activating Token Index: 389

Text #9

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0551. Min Act: -0.0001

Data Index: 421453 (C4 (Web Text))

Max Activating Token Index: 219

Text #10

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0550. Min Act: -0.0001

Data Index: 1488266 (Python Code)

Max Activating Token Index: 791

Text #11

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0478. Min Act: -0.0001

Data Index: 1187259 (C4 (Web Text))

Max Activating Token Index: 214

Text #12

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0585. Min Act: -0.0001

Data Index: 163061 (C4 (Web Text))

Max Activating Token Index: 600

Text #13

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0571. Min Act: -0.0001

Data Index: 1036177 (C4 (Web Text))

Max Activating Token Index: 389

Text #14

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0505. Min Act: -0.0001

Data Index: 1443070 (Python Code)

Max Activating Token Index: 708

Text #15

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0565. Min Act: -0.0001

Data Index: 858526 (C4 (Web Text))

Max Activating Token Index: 14

Text #16

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0503. Min Act: -0.0001

Data Index: 212567 (C4 (Web Text))

Max Activating Token Index: 159

Text #17

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0590. Min Act: -0.0001

Data Index: 858248 (C4 (Web Text))

Max Activating Token Index: 643

Text #18

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0523. Min Act: -0.0001

Data Index: 426805 (C4 (Web Text))

Max Activating Token Index: 561

Text #19

Max Range: 0.0705. Min Range: -0.0705

Max Act: 0.0543. Min Act: -0.0001

Data Index: 453181 (C4 (Web Text))

Max Activating Token Index: 490