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

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

Neuron 2677 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: 0.8863. Min Range: -0.8863

Max Act: 0.8863. Min Act: -0.0001

Data Index: 754464 (C4 (Web Text))

Max Activating Token Index: 406

Text #1

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.7520. Min Act: -0.0001

Data Index: 529196 (C4 (Web Text))

Max Activating Token Index: 681

Text #2

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.6659. Min Act: -0.0001

Data Index: 30164 (C4 (Web Text))

Max Activating Token Index: 18

Text #3

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5852. Min Act: -0.0001

Data Index: 650355 (C4 (Web Text))

Max Activating Token Index: 133

Text #4

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5405. Min Act: -0.0001

Data Index: 679655 (C4 (Web Text))

Max Activating Token Index: 453

Text #5

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5261. Min Act: -0.0001

Data Index: 1482760 (Python Code)

Max Activating Token Index: 478

Text #6

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5007. Min Act: -0.0001

Data Index: 1066957 (C4 (Web Text))

Max Activating Token Index: 743

Text #7

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5303. Min Act: -0.0001

Data Index: 332020 (C4 (Web Text))

Max Activating Token Index: 429

Text #8

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4889. Min Act: -0.0001

Data Index: 1410208 (Python Code)

Max Activating Token Index: 632

Text #9

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4810. Min Act: -0.0001

Data Index: 1275231 (C4 (Web Text))

Max Activating Token Index: 902

Text #10

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.5012. Min Act: -0.0001

Data Index: 736396 (C4 (Web Text))

Max Activating Token Index: 209

Text #11

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4544. Min Act: -0.0001

Data Index: 343505 (C4 (Web Text))

Max Activating Token Index: 753

Text #12

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4674. Min Act: -0.0001

Data Index: 338313 (C4 (Web Text))

Max Activating Token Index: 353

Text #13

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4596. Min Act: -0.0001

Data Index: 316940 (C4 (Web Text))

Max Activating Token Index: 504

Text #14

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4540. Min Act: -0.0001

Data Index: 605303 (C4 (Web Text))

Max Activating Token Index: 573

Text #15

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4681. Min Act: -0.0001

Data Index: 192871 (C4 (Web Text))

Max Activating Token Index: 30

Text #16

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4424. Min Act: -0.0001

Data Index: 892599 (C4 (Web Text))

Max Activating Token Index: 758

Text #17

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4293. Min Act: -0.0001

Data Index: 359021 (C4 (Web Text))

Max Activating Token Index: 223

Text #18

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4491. Min Act: -0.0001

Data Index: 418419 (C4 (Web Text))

Max Activating Token Index: 366

Text #19

Max Range: 0.8863. Min Range: -0.8863

Max Act: 0.4571. Min Act: -0.0001

Data Index: 1104082 (C4 (Web Text))

Max Activating Token Index: 308