Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 90 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.2642. Min Act: -0.1693

Data Index: 1420975 (Python Code)

Max Activating Token Index: 471

Text #1

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.2348. Min Act: -0.1700

Data Index: 1524727 (Python Code)

Max Activating Token Index: 404

Text #2

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.2325. Min Act: -0.1700

Data Index: 1457637 (Python Code)

Max Activating Token Index: 325

Text #3

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.2305. Min Act: -0.1700

Data Index: 1474176 (Python Code)

Max Activating Token Index: 307

Text #4

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1855. Min Act: -0.1699

Data Index: 1494973 (Python Code)

Max Activating Token Index: 565

Text #5

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1840. Min Act: -0.1700

Data Index: 1466327 (Python Code)

Max Activating Token Index: 374

Text #6

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1994. Min Act: -0.1700

Data Index: 1578257 (Python Code)

Max Activating Token Index: 501

Text #7

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1780. Min Act: -0.1699

Data Index: 1453532 (Python Code)

Max Activating Token Index: 540

Text #8

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1509. Min Act: -0.1666

Data Index: 1542506 (Python Code)

Max Activating Token Index: 988

Text #9

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1606. Min Act: -0.1699

Data Index: 1591674 (Python Code)

Max Activating Token Index: 260

Text #10

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1518. Min Act: -0.1698

Data Index: 1535649 (Python Code)

Max Activating Token Index: 145

Text #11

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1246. Min Act: -0.1480

Data Index: 1495909 (Python Code)

Max Activating Token Index: 900

Text #12

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1230. Min Act: -0.1647

Data Index: 1398522 (Python Code)

Max Activating Token Index: 439

Text #13

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1317. Min Act: -0.1690

Data Index: 1571203 (Python Code)

Max Activating Token Index: 259

Text #14

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1447. Min Act: -0.1693

Data Index: 1445868 (Python Code)

Max Activating Token Index: 272

Text #15

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.1370. Min Act: -0.1697

Data Index: 1456002 (Python Code)

Max Activating Token Index: 402

Text #16

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.0899. Min Act: -0.1700

Data Index: 1611228 (Python Code)

Max Activating Token Index: 249

Text #17

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.0855. Min Act: -0.1696

Data Index: 1488734 (Python Code)

Max Activating Token Index: 136

Text #18

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.0854. Min Act: -0.1653

Data Index: 1430596 (Python Code)

Max Activating Token Index: 877

Text #19

Max Range: 4.2642. Min Range: -4.2642

Max Act: 4.0957. Min Act: -0.1700

Data Index: 1568022 (Python Code)

Max Activating Token Index: 203