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

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

Neuron 617 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: 5.2368. Min Range: -5.2368

Max Act: 5.2368. Min Act: -0.1700

Data Index: 574152 (C4 (Web Text))

Max Activating Token Index: 322

Text #1

Max Range: 5.2368. Min Range: -5.2368

Max Act: 5.1431. Min Act: -0.1700

Data Index: 807411 (C4 (Web Text))

Max Activating Token Index: 962

Text #2

Max Range: 5.2368. Min Range: -5.2368

Max Act: 5.0806. Min Act: -0.1700

Data Index: 1040644 (C4 (Web Text))

Max Activating Token Index: 93

Text #3

Max Range: 5.2368. Min Range: -5.2368

Max Act: 5.1118. Min Act: -0.1700

Data Index: 650959 (C4 (Web Text))

Max Activating Token Index: 541

Text #4

Max Range: 5.2368. Min Range: -5.2368

Max Act: 5.0181. Min Act: -0.1700

Data Index: 996950 (C4 (Web Text))

Max Activating Token Index: 1001

Text #5

Max Range: 5.2368. Min Range: -5.2368

Max Act: 5.0181. Min Act: -0.1700

Data Index: 759132 (C4 (Web Text))

Max Activating Token Index: 538

Text #6

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9868. Min Act: -0.1700

Data Index: 232698 (C4 (Web Text))

Max Activating Token Index: 482

Text #7

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9868. Min Act: -0.1700

Data Index: 973236 (C4 (Web Text))

Max Activating Token Index: 849

Text #8

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9556. Min Act: -0.1700

Data Index: 378370 (C4 (Web Text))

Max Activating Token Index: 77

Text #9

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9243. Min Act: -0.1700

Data Index: 1294758 (C4 (Web Text))

Max Activating Token Index: 967

Text #10

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9556. Min Act: -0.1700

Data Index: 1242876 (C4 (Web Text))

Max Activating Token Index: 276

Text #11

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.9243. Min Act: -0.1700

Data Index: 746294 (C4 (Web Text))

Max Activating Token Index: 307

Text #12

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8931. Min Act: -0.1700

Data Index: 27816 (C4 (Web Text))

Max Activating Token Index: 149

Text #13

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8931. Min Act: -0.1699

Data Index: 89488 (C4 (Web Text))

Max Activating Token Index: 145

Text #14

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8618. Min Act: -0.1700

Data Index: 525517 (C4 (Web Text))

Max Activating Token Index: 544

Text #15

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8931. Min Act: -0.1700

Data Index: 194999 (C4 (Web Text))

Max Activating Token Index: 554

Text #16

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8618. Min Act: -0.1700

Data Index: 627750 (C4 (Web Text))

Max Activating Token Index: 1000

Text #17

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8618. Min Act: -0.1700

Data Index: 1011271 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8618. Min Act: -0.1700

Data Index: 785900 (C4 (Web Text))

Max Activating Token Index: 534

Text #19

Max Range: 5.2368. Min Range: -5.2368

Max Act: 4.8618. Min Act: -0.1700

Data Index: 915850 (C4 (Web Text))

Max Activating Token Index: 514