Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 311 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0573. Min Act: -0.0006

Data Index: 1283486 (C4 (Web Text))

Max Activating Token Index: 796

Text #1

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0563. Min Act: -0.0006

Data Index: 1385165 (Python Code)

Max Activating Token Index: 870

Text #2

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0565. Min Act: -0.0006

Data Index: 598209 (C4 (Web Text))

Max Activating Token Index: 1006

Text #3

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0562. Min Act: -0.0006

Data Index: 328881 (C4 (Web Text))

Max Activating Token Index: 779

Text #4

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0536. Min Act: -0.0006

Data Index: 653940 (C4 (Web Text))

Max Activating Token Index: 707

Text #5

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0543. Min Act: -0.0006

Data Index: 552599 (C4 (Web Text))

Max Activating Token Index: 328

Text #6

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0576. Min Act: -0.0006

Data Index: 426807 (C4 (Web Text))

Max Activating Token Index: 592

Text #7

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0541. Min Act: -0.0006

Data Index: 1256425 (C4 (Web Text))

Max Activating Token Index: 888

Text #8

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0549. Min Act: -0.0006

Data Index: 1632057 (Python Code)

Max Activating Token Index: 860

Text #9

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0578. Min Act: -0.0006

Data Index: 1363729 (Python Code)

Max Activating Token Index: 893

Text #10

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0521. Min Act: -0.0006

Data Index: 988772 (C4 (Web Text))

Max Activating Token Index: 831

Text #11

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0517. Min Act: -0.0006

Data Index: 774789 (C4 (Web Text))

Max Activating Token Index: 799

Text #12

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0532. Min Act: -0.0006

Data Index: 1312137 (C4 (Web Text))

Max Activating Token Index: 900

Text #13

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0504. Min Act: -0.0006

Data Index: 1515970 (Python Code)

Max Activating Token Index: 814

Text #14

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0514. Min Act: -0.0006

Data Index: 426387 (C4 (Web Text))

Max Activating Token Index: 605

Text #15

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0536. Min Act: -0.0006

Data Index: 1208400 (C4 (Web Text))

Max Activating Token Index: 593

Text #16

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0536. Min Act: -0.0006

Data Index: 882388 (C4 (Web Text))

Max Activating Token Index: 688

Text #17

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0532. Min Act: -0.0006

Data Index: 369596 (C4 (Web Text))

Max Activating Token Index: 938

Text #18

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0503. Min Act: -0.0006

Data Index: 412612 (C4 (Web Text))

Max Activating Token Index: 801

Text #19

Max Range: 0.0578. Min Range: -0.0578

Max Act: 0.0532. Min Act: -0.0006

Data Index: 1375149 (Python Code)

Max Activating Token Index: 848