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

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

Neuron 1533 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: 3.3339. Min Range: -3.3339

Max Act: 3.3339. Min Act: -0.1700

Data Index: 213225 (C4 (Web Text))

Max Activating Token Index: 650

Text #1

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.3339. Min Act: -0.1700

Data Index: 476019 (C4 (Web Text))

Max Activating Token Index: 281

Text #2

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.1924. Min Act: -0.1700

Data Index: 795378 (C4 (Web Text))

Max Activating Token Index: 128

Text #3

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.1610. Min Act: -0.1700

Data Index: 622574 (C4 (Web Text))

Max Activating Token Index: 323

Text #4

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.1295. Min Act: -0.1700

Data Index: 1250724 (C4 (Web Text))

Max Activating Token Index: 539

Text #5

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0822. Min Act: -0.1700

Data Index: 258398 (C4 (Web Text))

Max Activating Token Index: 655

Text #6

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0979. Min Act: -0.1700

Data Index: 354801 (C4 (Web Text))

Max Activating Token Index: 25

Text #7

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.1295. Min Act: -0.1700

Data Index: 228619 (C4 (Web Text))

Max Activating Token Index: 637

Text #8

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0979. Min Act: -0.1700

Data Index: 868823 (C4 (Web Text))

Max Activating Token Index: 724

Text #9

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.1137. Min Act: -0.1700

Data Index: 560410 (C4 (Web Text))

Max Activating Token Index: 1018

Text #10

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0979. Min Act: -0.1700

Data Index: 278365 (C4 (Web Text))

Max Activating Token Index: 658

Text #11

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0822. Min Act: -0.1700

Data Index: 640618 (C4 (Web Text))

Max Activating Token Index: 693

Text #12

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0506. Min Act: -0.1700

Data Index: 346628 (C4 (Web Text))

Max Activating Token Index: 878

Text #13

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0822. Min Act: -0.1700

Data Index: 1251069 (C4 (Web Text))

Max Activating Token Index: 669

Text #14

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0822. Min Act: -0.1700

Data Index: 291062 (C4 (Web Text))

Max Activating Token Index: 989

Text #15

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0506. Min Act: -0.1700

Data Index: 1484884 (Python Code)

Max Activating Token Index: 665

Text #16

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0190. Min Act: -0.1700

Data Index: 592931 (C4 (Web Text))

Max Activating Token Index: 537

Text #17

Max Range: 3.3339. Min Range: -3.3339

Max Act: 2.9874. Min Act: -0.1700

Data Index: 544355 (C4 (Web Text))

Max Activating Token Index: 885

Text #18

Max Range: 3.3339. Min Range: -3.3339

Max Act: 3.0348. Min Act: -0.1700

Data Index: 700662 (C4 (Web Text))

Max Activating Token Index: 750

Text #19

Max Range: 3.3339. Min Range: -3.3339

Max Act: 2.9874. Min Act: -0.1700

Data Index: 493432 (C4 (Web Text))

Max Activating Token Index: 812