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

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

Neuron 656 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0506. Min Range: -7.0506

Max Act: 7.0506. Min Act: -0.1700

Data Index: 1109729 (C4 (Web Text))

Max Activating Token Index: 478

Text #1

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8943. Min Act: -0.1700

Data Index: 441987 (C4 (Web Text))

Max Activating Token Index: 629

Text #2

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8943. Min Act: -0.1699

Data Index: 969361 (C4 (Web Text))

Max Activating Token Index: 575

Text #3

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8631. Min Act: -0.1700

Data Index: 38112 (C4 (Web Text))

Max Activating Token Index: 396

Text #4

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.9568. Min Act: -0.1700

Data Index: 1132327 (C4 (Web Text))

Max Activating Token Index: 555

Text #5

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.9256. Min Act: -0.1700

Data Index: 104348 (C4 (Web Text))

Max Activating Token Index: 683

Text #6

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8631. Min Act: -0.1700

Data Index: 558429 (C4 (Web Text))

Max Activating Token Index: 386

Text #7

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8943. Min Act: -0.1700

Data Index: 1230395 (C4 (Web Text))

Max Activating Token Index: 387

Text #8

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8943. Min Act: -0.1700

Data Index: 218274 (C4 (Web Text))

Max Activating Token Index: 532

Text #9

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8943. Min Act: -0.1700

Data Index: 406745 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.7381. Min Act: -0.1700

Data Index: 176984 (C4 (Web Text))

Max Activating Token Index: 440

Text #11

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8006. Min Act: -0.1700

Data Index: 306903 (C4 (Web Text))

Max Activating Token Index: 269

Text #12

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.8631. Min Act: -0.1692

Data Index: 361651 (C4 (Web Text))

Max Activating Token Index: 790

Text #13

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.6131. Min Act: -0.1700

Data Index: 922082 (C4 (Web Text))

Max Activating Token Index: 444

Text #14

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.5506. Min Act: -0.1700

Data Index: 1347580 (C4 (Web Text))

Max Activating Token Index: 843

Text #15

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.6443. Min Act: -0.1699

Data Index: 1078321 (C4 (Web Text))

Max Activating Token Index: 452

Text #16

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.6443. Min Act: -0.1700

Data Index: 1198123 (C4 (Web Text))

Max Activating Token Index: 270

Text #17

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.6131. Min Act: -0.1696

Data Index: 1024826 (C4 (Web Text))

Max Activating Token Index: 571

Text #18

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.6443. Min Act: -0.1700

Data Index: 391669 (C4 (Web Text))

Max Activating Token Index: 469

Text #19

Max Range: 7.0506. Min Range: -7.0506

Max Act: 6.5193. Min Act: -0.1700

Data Index: 987589 (C4 (Web Text))

Max Activating Token Index: 362