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

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

Neuron 2563 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1784. Min Range: -5.1784

Max Act: 5.1784. Min Act: -0.0001

Data Index: 790156 (C4 (Web Text))

Max Activating Token Index: 507

Text #1

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.9342. Min Act: -0.0001

Data Index: 1318237 (C4 (Web Text))

Max Activating Token Index: 773

Text #2

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.9286. Min Act: -0.0001

Data Index: 187319 (C4 (Web Text))

Max Activating Token Index: 428

Text #3

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.6563. Min Act: -0.0001

Data Index: 503755 (C4 (Web Text))

Max Activating Token Index: 438

Text #4

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.9743. Min Act: -0.0001

Data Index: 400630 (C4 (Web Text))

Max Activating Token Index: 287

Text #5

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.9715. Min Act: -0.0001

Data Index: 1014664 (C4 (Web Text))

Max Activating Token Index: 218

Text #6

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.4446. Min Act: -0.0001

Data Index: 748415 (C4 (Web Text))

Max Activating Token Index: 583

Text #7

Max Range: 5.1784. Min Range: -5.1784

Max Act: 5.1204. Min Act: -0.0001

Data Index: 333281 (C4 (Web Text))

Max Activating Token Index: 786

Text #8

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.5005. Min Act: -0.0001

Data Index: 1346713 (C4 (Web Text))

Max Activating Token Index: 271

Text #9

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.4876. Min Act: -0.0001

Data Index: 1217027 (C4 (Web Text))

Max Activating Token Index: 670

Text #10

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.6449. Min Act: -0.0001

Data Index: 413931 (C4 (Web Text))

Max Activating Token Index: 327

Text #11

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.4653. Min Act: -0.0001

Data Index: 180763 (C4 (Web Text))

Max Activating Token Index: 631

Text #12

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.6384. Min Act: -0.0001

Data Index: 547275 (C4 (Web Text))

Max Activating Token Index: 270

Text #13

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.6340. Min Act: -0.0001

Data Index: 520480 (C4 (Web Text))

Max Activating Token Index: 409

Text #14

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.7978. Min Act: -0.0001

Data Index: 604392 (C4 (Web Text))

Max Activating Token Index: 613

Text #15

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.7833. Min Act: -0.0001

Data Index: 106288 (C4 (Web Text))

Max Activating Token Index: 586

Text #16

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.5775. Min Act: -0.0001

Data Index: 707467 (C4 (Web Text))

Max Activating Token Index: 851

Text #17

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.3486. Min Act: -0.0001

Data Index: 515726 (C4 (Web Text))

Max Activating Token Index: 712

Text #18

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.5006. Min Act: -0.0001

Data Index: 1066327 (C4 (Web Text))

Max Activating Token Index: 314

Text #19

Max Range: 5.1784. Min Range: -5.1784

Max Act: 4.6766. Min Act: -0.0001

Data Index: 1188277 (C4 (Web Text))

Max Activating Token Index: 685