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

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

Neuron 1022 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.7385. Min Act: -0.1700

Data Index: 1438950 (Python Code)

Max Activating Token Index: 600

Text #1

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.4885. Min Act: -0.1700

Data Index: 1581795 (Python Code)

Max Activating Token Index: 745

Text #2

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.4260. Min Act: -0.1700

Data Index: 1447395 (Python Code)

Max Activating Token Index: 635

Text #3

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.3635. Min Act: -0.1700

Data Index: 1605415 (Python Code)

Max Activating Token Index: 506

Text #4

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.3635. Min Act: -0.1700

Data Index: 1360731 (Python Code)

Max Activating Token Index: 488

Text #5

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.3010. Min Act: -0.1700

Data Index: 1386477 (Python Code)

Max Activating Token Index: 812

Text #6

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.2385. Min Act: -0.1700

Data Index: 1464639 (Python Code)

Max Activating Token Index: 311

Text #7

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.1760. Min Act: -0.1700

Data Index: 1645871 (Python Code)

Max Activating Token Index: 984

Text #8

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.1760. Min Act: -0.1700

Data Index: 1454271 (Python Code)

Max Activating Token Index: 772

Text #9

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.1135. Min Act: -0.1700

Data Index: 1446114 (Python Code)

Max Activating Token Index: 141

Text #10

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.0510. Min Act: -0.1700

Data Index: 1411340 (Python Code)

Max Activating Token Index: 1005

Text #11

Max Range: 8.7385. Min Range: -8.7385

Max Act: 8.0510. Min Act: -0.1700

Data Index: 1387862 (Python Code)

Max Activating Token Index: 1010

Text #12

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.9885. Min Act: -0.1700

Data Index: 1483633 (Python Code)

Max Activating Token Index: 960

Text #13

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.9885. Min Act: -0.1700

Data Index: 1610965 (Python Code)

Max Activating Token Index: 445

Text #14

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.8010. Min Act: -0.1700

Data Index: 1574329 (Python Code)

Max Activating Token Index: 969

Text #15

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.9885. Min Act: -0.1700

Data Index: 1443058 (Python Code)

Max Activating Token Index: 1018

Text #16

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.9260. Min Act: -0.1700

Data Index: 1518183 (Python Code)

Max Activating Token Index: 421

Text #17

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.9260. Min Act: -0.1700

Data Index: 1582672 (Python Code)

Max Activating Token Index: 645

Text #18

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.8635. Min Act: -0.1700

Data Index: 1540077 (Python Code)

Max Activating Token Index: 393

Text #19

Max Range: 8.7385. Min Range: -8.7385

Max Act: 7.8635. Min Act: -0.1700

Data Index: 1482048 (Python Code)

Max Activating Token Index: 968