Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 359 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.2694. Min Act: -0.1700

Data Index: 1548721 (Python Code)

Max Activating Token Index: 737

Text #1

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.2381. Min Act: -0.1700

Data Index: 1477576 (Python Code)

Max Activating Token Index: 998

Text #2

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.1444. Min Act: -0.1700

Data Index: 1437573 (Python Code)

Max Activating Token Index: 161

Text #3

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0819. Min Act: -0.1700

Data Index: 1519805 (Python Code)

Max Activating Token Index: 653

Text #4

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0975. Min Act: -0.1700

Data Index: 1371683 (Python Code)

Max Activating Token Index: 365

Text #5

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0662. Min Act: -0.1700

Data Index: 1588832 (Python Code)

Max Activating Token Index: 835

Text #6

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0975. Min Act: -0.1700

Data Index: 1375185 (Python Code)

Max Activating Token Index: 259

Text #7

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0819. Min Act: -0.1700

Data Index: 1384544 (Python Code)

Max Activating Token Index: 787

Text #8

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0350. Min Act: -0.1700

Data Index: 1652155 (Python Code)

Max Activating Token Index: 784

Text #9

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0506. Min Act: -0.1700

Data Index: 1530668 (Python Code)

Max Activating Token Index: 515

Text #10

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0194. Min Act: -0.1700

Data Index: 1584787 (Python Code)

Max Activating Token Index: 460

Text #11

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.9881. Min Act: -0.1700

Data Index: 1505852 (Python Code)

Max Activating Token Index: 109

Text #12

Max Range: 6.2694. Min Range: -6.2694

Max Act: 6.0194. Min Act: -0.1699

Data Index: 1499759 (Python Code)

Max Activating Token Index: 397

Text #13

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.9569. Min Act: -0.1700

Data Index: 1492071 (Python Code)

Max Activating Token Index: 798

Text #14

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.8319. Min Act: -0.1700

Data Index: 1452821 (Python Code)

Max Activating Token Index: 696

Text #15

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.9412. Min Act: -0.1699

Data Index: 1396769 (Python Code)

Max Activating Token Index: 758

Text #16

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.9412. Min Act: -0.1700

Data Index: 1408657 (Python Code)

Max Activating Token Index: 481

Text #17

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.9256. Min Act: -0.1700

Data Index: 1477713 (Python Code)

Max Activating Token Index: 304

Text #18

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.8631. Min Act: -0.1700

Data Index: 1398856 (Python Code)

Max Activating Token Index: 183

Text #19

Max Range: 6.2694. Min Range: -6.2694

Max Act: 5.8319. Min Act: -0.1700

Data Index: 1507788 (Python Code)

Max Activating Token Index: 1007