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

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

Neuron 274 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.4217. Min Act: -0.0000

Data Index: 1345436 (C4 (Web Text))

Max Activating Token Index: 707

Text #1

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.4341. Min Act: -0.0000

Data Index: 1039394 (C4 (Web Text))

Max Activating Token Index: 40

Text #2

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.4076. Min Act: -0.0000

Data Index: 1452045 (Python Code)

Max Activating Token Index: 147

Text #3

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.4188. Min Act: -0.0001

Data Index: 1411749 (Python Code)

Max Activating Token Index: 973

Text #4

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3713. Min Act: -0.0000

Data Index: 771088 (C4 (Web Text))

Max Activating Token Index: 22

Text #5

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3263. Min Act: -0.0000

Data Index: 120635 (C4 (Web Text))

Max Activating Token Index: 801

Text #6

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3946. Min Act: -0.0000

Data Index: 95616 (C4 (Web Text))

Max Activating Token Index: 334

Text #7

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3393. Min Act: -0.0001

Data Index: 343651 (C4 (Web Text))

Max Activating Token Index: 312

Text #8

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3552. Min Act: -0.0000

Data Index: 493812 (C4 (Web Text))

Max Activating Token Index: 511

Text #9

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3680. Min Act: -0.0001

Data Index: 88175 (C4 (Web Text))

Max Activating Token Index: 578

Text #10

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3326. Min Act: -0.0001

Data Index: 220458 (C4 (Web Text))

Max Activating Token Index: 289

Text #11

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3417. Min Act: -0.0001

Data Index: 1312541 (C4 (Web Text))

Max Activating Token Index: 454

Text #12

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3378. Min Act: -0.0000

Data Index: 173414 (C4 (Web Text))

Max Activating Token Index: 368

Text #13

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3627. Min Act: -0.0001

Data Index: 1446402 (Python Code)

Max Activating Token Index: 868

Text #14

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3005. Min Act: -0.0001

Data Index: 74703 (C4 (Web Text))

Max Activating Token Index: 276

Text #15

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3297. Min Act: -0.0001

Data Index: 410162 (C4 (Web Text))

Max Activating Token Index: 424

Text #16

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.2825. Min Act: -0.0000

Data Index: 870493 (C4 (Web Text))

Max Activating Token Index: 143

Text #17

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3311. Min Act: -0.0000

Data Index: 196341 (C4 (Web Text))

Max Activating Token Index: 420

Text #18

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3407. Min Act: -0.0001

Data Index: 381747 (C4 (Web Text))

Max Activating Token Index: 267

Text #19

Max Range: 0.4341. Min Range: -0.4341

Max Act: 0.3324. Min Act: -0.0000

Data Index: 471735 (C4 (Web Text))

Max Activating Token Index: 568