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

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

Neuron 1501 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: 7.2250. Min Range: -7.2250

Max Act: 7.2250. Min Act: -0.1700

Data Index: 1497864 (Python Code)

Max Activating Token Index: 1002

Text #1

Max Range: 7.2250. Min Range: -7.2250

Max Act: 7.0688. Min Act: -0.1700

Data Index: 1364343 (Python Code)

Max Activating Token Index: 321

Text #2

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.9750. Min Act: -0.1700

Data Index: 1542139 (Python Code)

Max Activating Token Index: 478

Text #3

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.9438. Min Act: -0.1700

Data Index: 1520243 (Python Code)

Max Activating Token Index: 356

Text #4

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.8813. Min Act: -0.1693

Data Index: 1448211 (Python Code)

Max Activating Token Index: 303

Text #5

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.8188. Min Act: -0.1700

Data Index: 1499914 (Python Code)

Max Activating Token Index: 287

Text #6

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.7875. Min Act: -0.1700

Data Index: 1423370 (Python Code)

Max Activating Token Index: 806

Text #7

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.7875. Min Act: -0.1700

Data Index: 1476880 (Python Code)

Max Activating Token Index: 51

Text #8

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6625. Min Act: -0.1700

Data Index: 1398405 (Python Code)

Max Activating Token Index: 760

Text #9

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6313. Min Act: -0.1700

Data Index: 1510498 (Python Code)

Max Activating Token Index: 328

Text #10

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6313. Min Act: -0.1699

Data Index: 1401801 (Python Code)

Max Activating Token Index: 321

Text #11

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6313. Min Act: -0.1686

Data Index: 1500741 (Python Code)

Max Activating Token Index: 500

Text #12

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6938. Min Act: -0.1700

Data Index: 1648292 (Python Code)

Max Activating Token Index: 929

Text #13

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.5375. Min Act: -0.1700

Data Index: 1593798 (Python Code)

Max Activating Token Index: 225

Text #14

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6000. Min Act: -0.1700

Data Index: 1437535 (Python Code)

Max Activating Token Index: 391

Text #15

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.5688. Min Act: -0.1700

Data Index: 1588282 (Python Code)

Max Activating Token Index: 282

Text #16

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.5688. Min Act: -0.1700

Data Index: 1566481 (Python Code)

Max Activating Token Index: 530

Text #17

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.5688. Min Act: -0.1700

Data Index: 1422678 (Python Code)

Max Activating Token Index: 402

Text #18

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.5688. Min Act: -0.1668

Data Index: 1359870 (Python Code)

Max Activating Token Index: 955

Text #19

Max Range: 7.2250. Min Range: -7.2250

Max Act: 6.6000. Min Act: -0.1689

Data Index: 1562582 (Python Code)

Max Activating Token Index: 971