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

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

Neuron 152 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: 3.4502. Min Range: -3.4502

Max Act: 3.4446. Min Act: -0.1596

Data Index: 1428478 (Python Code)

Max Activating Token Index: 493

Text #1

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.4502. Min Act: -0.1699

Data Index: 1413893 (Python Code)

Max Activating Token Index: 415

Text #2

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.3757. Min Act: -0.1700

Data Index: 1474523 (Python Code)

Max Activating Token Index: 74

Text #3

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.3647. Min Act: -0.1698

Data Index: 1539782 (Python Code)

Max Activating Token Index: 966

Text #4

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.3747. Min Act: -0.1674

Data Index: 1487966 (Python Code)

Max Activating Token Index: 955

Text #5

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.3374. Min Act: -0.1699

Data Index: 305276 (C4 (Web Text))

Max Activating Token Index: 305

Text #6

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.2795. Min Act: -0.1699

Data Index: 1384302 (Python Code)

Max Activating Token Index: 864

Text #7

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.2584. Min Act: -0.1509

Data Index: 1599098 (Python Code)

Max Activating Token Index: 43

Text #8

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.2005. Min Act: -0.1052

Data Index: 1655913 (Python Code)

Max Activating Token Index: 927

Text #9

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1955. Min Act: -0.1699

Data Index: 1635678 (Python Code)

Max Activating Token Index: 52

Text #10

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1460. Min Act: -0.1699

Data Index: 332945 (C4 (Web Text))

Max Activating Token Index: 125

Text #11

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1358. Min Act: -0.1700

Data Index: 1568537 (Python Code)

Max Activating Token Index: 519

Text #12

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1511. Min Act: -0.1700

Data Index: 1008544 (C4 (Web Text))

Max Activating Token Index: 385

Text #13

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1383. Min Act: -0.1699

Data Index: 1614103 (Python Code)

Max Activating Token Index: 278

Text #14

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1391. Min Act: -0.1700

Data Index: 140199 (C4 (Web Text))

Max Activating Token Index: 990

Text #15

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1172. Min Act: -0.1682

Data Index: 1371276 (Python Code)

Max Activating Token Index: 797

Text #16

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1127. Min Act: -0.1700

Data Index: 1448814 (Python Code)

Max Activating Token Index: 390

Text #17

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1199. Min Act: -0.1700

Data Index: 1448967 (Python Code)

Max Activating Token Index: 1009

Text #18

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1190. Min Act: -0.1699

Data Index: 131258 (C4 (Web Text))

Max Activating Token Index: 584

Text #19

Max Range: 3.4502. Min Range: -3.4502

Max Act: 3.1103. Min Act: -0.1700

Data Index: 606867 (C4 (Web Text))

Max Activating Token Index: 888