Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 250 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0517. Min Act: -0.0002

Data Index: 506396 (C4 (Web Text))

Max Activating Token Index: 733

Text #1

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0459. Min Act: -0.0002

Data Index: 816501 (C4 (Web Text))

Max Activating Token Index: 772

Text #2

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0434. Min Act: -0.0002

Data Index: 319176 (C4 (Web Text))

Max Activating Token Index: 869

Text #3

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0432. Min Act: -0.0003

Data Index: 1149339 (C4 (Web Text))

Max Activating Token Index: 583

Text #4

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0410. Min Act: -0.0002

Data Index: 889501 (C4 (Web Text))

Max Activating Token Index: 632

Text #5

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0382. Min Act: -0.0002

Data Index: 825250 (C4 (Web Text))

Max Activating Token Index: 242

Text #6

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0379. Min Act: -0.0003

Data Index: 448360 (C4 (Web Text))

Max Activating Token Index: 300

Text #7

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0373. Min Act: -0.0002

Data Index: 1138407 (C4 (Web Text))

Max Activating Token Index: 478

Text #8

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0370. Min Act: -0.0002

Data Index: 1318092 (C4 (Web Text))

Max Activating Token Index: 780

Text #9

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0360. Min Act: -0.0002

Data Index: 884161 (C4 (Web Text))

Max Activating Token Index: 903

Text #10

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0344. Min Act: -0.0002

Data Index: 1129506 (C4 (Web Text))

Max Activating Token Index: 125

Text #11

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0356. Min Act: -0.0003

Data Index: 941685 (C4 (Web Text))

Max Activating Token Index: 246

Text #12

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0353. Min Act: -0.0002

Data Index: 1559603 (Python Code)

Max Activating Token Index: 148

Text #13

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0346. Min Act: -0.0002

Data Index: 235845 (C4 (Web Text))

Max Activating Token Index: 202

Text #14

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0332. Min Act: -0.0003

Data Index: 403903 (C4 (Web Text))

Max Activating Token Index: 724

Text #15

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0345. Min Act: -0.0002

Data Index: 713909 (C4 (Web Text))

Max Activating Token Index: 572

Text #16

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0332. Min Act: -0.0003

Data Index: 85927 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0357. Min Act: -0.0002

Data Index: 551552 (C4 (Web Text))

Max Activating Token Index: 173

Text #18

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0343. Min Act: -0.0003

Data Index: 542199 (C4 (Web Text))

Max Activating Token Index: 507

Text #19

Max Range: 0.0517. Min Range: -0.0517

Max Act: 0.0341. Min Act: -0.0002

Data Index: 492257 (C4 (Web Text))

Max Activating Token Index: 792