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

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

Neuron 633 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.6339. Min Act: -0.1700

Data Index: 65205 (C4 (Web Text))

Max Activating Token Index: 290

Text #1

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5401. Min Act: -0.1700

Data Index: 1045266 (C4 (Web Text))

Max Activating Token Index: 659

Text #2

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5401. Min Act: -0.1700

Data Index: 1417039 (Python Code)

Max Activating Token Index: 656

Text #3

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.6026. Min Act: -0.1700

Data Index: 561742 (C4 (Web Text))

Max Activating Token Index: 791

Text #4

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5714. Min Act: -0.1700

Data Index: 85141 (C4 (Web Text))

Max Activating Token Index: 409

Text #5

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5401. Min Act: -0.1700

Data Index: 1425672 (Python Code)

Max Activating Token Index: 723

Text #6

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5089. Min Act: -0.1700

Data Index: 1438563 (Python Code)

Max Activating Token Index: 712

Text #7

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4776. Min Act: -0.1700

Data Index: 1642928 (Python Code)

Max Activating Token Index: 775

Text #8

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1523427 (Python Code)

Max Activating Token Index: 723

Text #9

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1376281 (Python Code)

Max Activating Token Index: 695

Text #10

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1514574 (Python Code)

Max Activating Token Index: 718

Text #11

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.5089. Min Act: -0.1700

Data Index: 1511289 (Python Code)

Max Activating Token Index: 846

Text #12

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1508705 (Python Code)

Max Activating Token Index: 638

Text #13

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.3839. Min Act: -0.1700

Data Index: 1626077 (Python Code)

Max Activating Token Index: 703

Text #14

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1450249 (Python Code)

Max Activating Token Index: 689

Text #15

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4151. Min Act: -0.1700

Data Index: 1521140 (Python Code)

Max Activating Token Index: 900

Text #16

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.3526. Min Act: -0.1700

Data Index: 1460195 (Python Code)

Max Activating Token Index: 763

Text #17

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4151. Min Act: -0.1700

Data Index: 1468849 (Python Code)

Max Activating Token Index: 849

Text #18

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4151. Min Act: -0.1700

Data Index: 1391814 (Python Code)

Max Activating Token Index: 783

Text #19

Max Range: 4.6339. Min Range: -4.6339

Max Act: 4.4464. Min Act: -0.1700

Data Index: 1443850 (Python Code)

Max Activating Token Index: 556