Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1603 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1642. Min Act: -0.0003

Data Index: 69666 (C4 (Web Text))

Max Activating Token Index: 716

Text #1

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1624. Min Act: -0.0003

Data Index: 818050 (C4 (Web Text))

Max Activating Token Index: 209

Text #2

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1722. Min Act: -0.0003

Data Index: 996472 (C4 (Web Text))

Max Activating Token Index: 415

Text #3

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1582. Min Act: -0.0003

Data Index: 1329394 (C4 (Web Text))

Max Activating Token Index: 444

Text #4

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1615. Min Act: -0.0003

Data Index: 555238 (C4 (Web Text))

Max Activating Token Index: 412

Text #5

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1633. Min Act: -0.0003

Data Index: 520701 (C4 (Web Text))

Max Activating Token Index: 240

Text #6

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1596. Min Act: -0.0002

Data Index: 10228 (C4 (Web Text))

Max Activating Token Index: 432

Text #7

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1741. Min Act: -0.0004

Data Index: 1299116 (C4 (Web Text))

Max Activating Token Index: 808

Text #8

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1578. Min Act: -0.0003

Data Index: 519186 (C4 (Web Text))

Max Activating Token Index: 242

Text #9

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1540. Min Act: -0.0003

Data Index: 36878 (C4 (Web Text))

Max Activating Token Index: 204

Text #10

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1615. Min Act: -0.0003

Data Index: 864303 (C4 (Web Text))

Max Activating Token Index: 235

Text #11

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1582. Min Act: -0.0003

Data Index: 1327127 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1516. Min Act: -0.0003

Data Index: 1266094 (C4 (Web Text))

Max Activating Token Index: 428

Text #13

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1531. Min Act: -0.0004

Data Index: 671458 (C4 (Web Text))

Max Activating Token Index: 414

Text #14

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1653. Min Act: -0.0004

Data Index: 543590 (C4 (Web Text))

Max Activating Token Index: 576

Text #15

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1573. Min Act: -0.0003

Data Index: 65494 (C4 (Web Text))

Max Activating Token Index: 722

Text #16

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1452. Min Act: -0.0003

Data Index: 1078671 (C4 (Web Text))

Max Activating Token Index: 202

Text #17

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1489. Min Act: -0.0003

Data Index: 1339838 (C4 (Web Text))

Max Activating Token Index: 501

Text #18

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1333. Min Act: -0.0003

Data Index: 266612 (C4 (Web Text))

Max Activating Token Index: 781

Text #19

Max Range: 0.1741. Min Range: -0.1741

Max Act: 0.1579. Min Act: -0.0003

Data Index: 1287978 (C4 (Web Text))

Max Activating Token Index: 866