Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2333 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1694. Min Act: -0.0001

Data Index: 588261 (C4 (Web Text))

Max Activating Token Index: 537

Text #1

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1735. Min Act: -0.0001

Data Index: 173295 (C4 (Web Text))

Max Activating Token Index: 666

Text #2

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1643. Min Act: -0.0001

Data Index: 202135 (C4 (Web Text))

Max Activating Token Index: 865

Text #3

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1646. Min Act: -0.0001

Data Index: 75626 (C4 (Web Text))

Max Activating Token Index: 362

Text #4

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1595. Min Act: -0.0001

Data Index: 1468551 (Python Code)

Max Activating Token Index: 16

Text #5

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1590. Min Act: -0.0001

Data Index: 411870 (C4 (Web Text))

Max Activating Token Index: 319

Text #6

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1536. Min Act: -0.0001

Data Index: 1256342 (C4 (Web Text))

Max Activating Token Index: 614

Text #7

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1547. Min Act: -0.0001

Data Index: 66900 (C4 (Web Text))

Max Activating Token Index: 503

Text #8

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1529. Min Act: -0.0001

Data Index: 640801 (C4 (Web Text))

Max Activating Token Index: 884

Text #9

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1522. Min Act: -0.0001

Data Index: 552928 (C4 (Web Text))

Max Activating Token Index: 225

Text #10

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1476. Min Act: -0.0001

Data Index: 289122 (C4 (Web Text))

Max Activating Token Index: 860

Text #11

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1438. Min Act: -0.0001

Data Index: 313303 (C4 (Web Text))

Max Activating Token Index: 270

Text #12

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1446. Min Act: -0.0001

Data Index: 918663 (C4 (Web Text))

Max Activating Token Index: 842

Text #13

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1456. Min Act: -0.0001

Data Index: 1053671 (C4 (Web Text))

Max Activating Token Index: 402

Text #14

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1477. Min Act: -0.0001

Data Index: 1205272 (C4 (Web Text))

Max Activating Token Index: 940

Text #15

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1408. Min Act: -0.0001

Data Index: 1036207 (C4 (Web Text))

Max Activating Token Index: 272

Text #16

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1423. Min Act: -0.0001

Data Index: 192381 (C4 (Web Text))

Max Activating Token Index: 458

Text #17

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1371. Min Act: -0.0001

Data Index: 1134859 (C4 (Web Text))

Max Activating Token Index: 216

Text #18

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1368. Min Act: -0.0001

Data Index: 224529 (C4 (Web Text))

Max Activating Token Index: 611

Text #19

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1420. Min Act: -0.0001

Data Index: 833004 (C4 (Web Text))

Max Activating Token Index: 974