Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3919 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.9477. Min Act: -0.0000

Data Index: 370117 (C4 (Web Text))

Max Activating Token Index: 199

Text #1

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.6488. Min Act: -0.0001

Data Index: 643509 (C4 (Web Text))

Max Activating Token Index: 18

Text #2

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5794. Min Act: -0.0001

Data Index: 770676 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5072. Min Act: -0.0001

Data Index: 1128950 (C4 (Web Text))

Max Activating Token Index: 58

Text #4

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5482. Min Act: -0.0001

Data Index: 1349176 (C4 (Web Text))

Max Activating Token Index: 89

Text #5

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5448. Min Act: -0.0001

Data Index: 1020774 (C4 (Web Text))

Max Activating Token Index: 538

Text #6

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4356. Min Act: -0.0001

Data Index: 991375 (C4 (Web Text))

Max Activating Token Index: 671

Text #7

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5188. Min Act: -0.0001

Data Index: 1126603 (C4 (Web Text))

Max Activating Token Index: 875

Text #8

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.5014. Min Act: -0.0001

Data Index: 717893 (C4 (Web Text))

Max Activating Token Index: 189

Text #9

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4830. Min Act: -0.0001

Data Index: 874814 (C4 (Web Text))

Max Activating Token Index: 469

Text #10

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4762. Min Act: -0.0000

Data Index: 836606 (C4 (Web Text))

Max Activating Token Index: 204

Text #11

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4765. Min Act: -0.0000

Data Index: 990382 (C4 (Web Text))

Max Activating Token Index: 507

Text #12

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4479. Min Act: -0.0001

Data Index: 543686 (C4 (Web Text))

Max Activating Token Index: 761

Text #13

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4605. Min Act: -0.0001

Data Index: 368629 (C4 (Web Text))

Max Activating Token Index: 927

Text #14

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4807. Min Act: -0.0000

Data Index: 1012787 (C4 (Web Text))

Max Activating Token Index: 1001

Text #15

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4452. Min Act: -0.0001

Data Index: 718553 (C4 (Web Text))

Max Activating Token Index: 361

Text #16

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.3734. Min Act: -0.0001

Data Index: 1370562 (Python Code)

Max Activating Token Index: 237

Text #17

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4371. Min Act: -0.0001

Data Index: 901762 (C4 (Web Text))

Max Activating Token Index: 301

Text #18

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4535. Min Act: -0.0000

Data Index: 308438 (C4 (Web Text))

Max Activating Token Index: 917

Text #19

Max Range: 0.9477. Min Range: -0.9477

Max Act: 0.4132. Min Act: -0.0001

Data Index: 1050329 (C4 (Web Text))

Max Activating Token Index: 516