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

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

Neuron 4208 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9582. Min Act: -0.0000

Data Index: 1243686 (C4 (Web Text))

Max Activating Token Index: 277

Text #1

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9984. Min Act: -0.0000

Data Index: 1479713 (Python Code)

Max Activating Token Index: 464

Text #2

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8983. Min Act: -0.0000

Data Index: 279157 (C4 (Web Text))

Max Activating Token Index: 403

Text #3

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8874. Min Act: -0.0000

Data Index: 70006 (C4 (Web Text))

Max Activating Token Index: 764

Text #4

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8918. Min Act: -0.0000

Data Index: 1043545 (C4 (Web Text))

Max Activating Token Index: 148

Text #5

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9845. Min Act: -0.0000

Data Index: 800188 (C4 (Web Text))

Max Activating Token Index: 617

Text #6

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8572. Min Act: -0.0000

Data Index: 13219 (C4 (Web Text))

Max Activating Token Index: 117

Text #7

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9481. Min Act: -0.0000

Data Index: 704975 (C4 (Web Text))

Max Activating Token Index: 398

Text #8

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9263. Min Act: 0.0000

Data Index: 1466944 (Python Code)

Max Activating Token Index: 397

Text #9

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9502. Min Act: -0.0000

Data Index: 1457293 (Python Code)

Max Activating Token Index: 860

Text #10

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8852. Min Act: -0.0000

Data Index: 584198 (C4 (Web Text))

Max Activating Token Index: 869

Text #11

Max Range: 1.0003. Min Range: -1.0003

Max Act: 1.0003. Min Act: -0.0000

Data Index: 1282412 (C4 (Web Text))

Max Activating Token Index: 932

Text #12

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9680. Min Act: -0.0000

Data Index: 1237370 (C4 (Web Text))

Max Activating Token Index: 523

Text #13

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8770. Min Act: -0.0000

Data Index: 207200 (C4 (Web Text))

Max Activating Token Index: 714

Text #14

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.9925. Min Act: -0.0000

Data Index: 81352 (C4 (Web Text))

Max Activating Token Index: 653

Text #15

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8901. Min Act: 0.0000

Data Index: 1532677 (Python Code)

Max Activating Token Index: 35

Text #16

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8358. Min Act: -0.0000

Data Index: 1320444 (C4 (Web Text))

Max Activating Token Index: 581

Text #17

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8848. Min Act: -0.0000

Data Index: 74326 (C4 (Web Text))

Max Activating Token Index: 457

Text #18

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8873. Min Act: -0.0000

Data Index: 436451 (C4 (Web Text))

Max Activating Token Index: 378

Text #19

Max Range: 1.0003. Min Range: -1.0003

Max Act: 0.8881. Min Act: 0.0001

Data Index: 1494858 (Python Code)

Max Activating Token Index: 740