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

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

Neuron 151 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: 8.2618. Min Range: -8.2618

Max Act: 8.2618. Min Act: -0.1700

Data Index: 20801 (C4 (Web Text))

Max Activating Token Index: 398

Text #1

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.2305. Min Act: -0.1700

Data Index: 159859 (C4 (Web Text))

Max Activating Token Index: 346

Text #2

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.2305. Min Act: -0.1700

Data Index: 444783 (C4 (Web Text))

Max Activating Token Index: 729

Text #3

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.1680. Min Act: -0.1700

Data Index: 431412 (C4 (Web Text))

Max Activating Token Index: 642

Text #4

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.1368. Min Act: -0.1700

Data Index: 1352630 (C4 (Web Text))

Max Activating Token Index: 556

Text #5

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.1680. Min Act: -0.1700

Data Index: 931009 (C4 (Web Text))

Max Activating Token Index: 718

Text #6

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.0743. Min Act: -0.1699

Data Index: 929472 (C4 (Web Text))

Max Activating Token Index: 437

Text #7

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.9805. Min Act: -0.1700

Data Index: 657624 (C4 (Web Text))

Max Activating Token Index: 84

Text #8

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.0118. Min Act: -0.1700

Data Index: 469309 (C4 (Web Text))

Max Activating Token Index: 838

Text #9

Max Range: 8.2618. Min Range: -8.2618

Max Act: 8.0430. Min Act: -0.1699

Data Index: 1135064 (C4 (Web Text))

Max Activating Token Index: 511

Text #10

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.9493. Min Act: -0.1700

Data Index: 796742 (C4 (Web Text))

Max Activating Token Index: 537

Text #11

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.9493. Min Act: -0.1700

Data Index: 509773 (C4 (Web Text))

Max Activating Token Index: 968

Text #12

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8555. Min Act: -0.1700

Data Index: 1218323 (C4 (Web Text))

Max Activating Token Index: 373

Text #13

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.9493. Min Act: -0.1700

Data Index: 254087 (C4 (Web Text))

Max Activating Token Index: 739

Text #14

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8243. Min Act: -0.1699

Data Index: 657214 (C4 (Web Text))

Max Activating Token Index: 1012

Text #15

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8555. Min Act: -0.1700

Data Index: 513501 (C4 (Web Text))

Max Activating Token Index: 665

Text #16

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8243. Min Act: -0.1700

Data Index: 339455 (C4 (Web Text))

Max Activating Token Index: 198

Text #17

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8555. Min Act: -0.1700

Data Index: 173616 (C4 (Web Text))

Max Activating Token Index: 338

Text #18

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.8243. Min Act: -0.1700

Data Index: 1153166 (C4 (Web Text))

Max Activating Token Index: 727

Text #19

Max Range: 8.2618. Min Range: -8.2618

Max Act: 7.7930. Min Act: -0.1700

Data Index: 122260 (C4 (Web Text))

Max Activating Token Index: 928