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

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

Neuron 272 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: 5.6763. Min Range: -5.6763

Max Act: 5.6763. Min Act: -0.1700

Data Index: 230266 (C4 (Web Text))

Max Activating Token Index: 736

Text #1

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.2700. Min Act: -0.1700

Data Index: 168090 (C4 (Web Text))

Max Activating Token Index: 938

Text #2

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.4888. Min Act: -0.1700

Data Index: 1335825 (C4 (Web Text))

Max Activating Token Index: 825

Text #3

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.3013. Min Act: -0.1700

Data Index: 586853 (C4 (Web Text))

Max Activating Token Index: 655

Text #4

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.1763. Min Act: -0.1699

Data Index: 97861 (C4 (Web Text))

Max Activating Token Index: 675

Text #5

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0825. Min Act: -0.1700

Data Index: 1081616 (C4 (Web Text))

Max Activating Token Index: 584

Text #6

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0825. Min Act: -0.1700

Data Index: 782518 (C4 (Web Text))

Max Activating Token Index: 279

Text #7

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0200. Min Act: -0.1700

Data Index: 1013539 (C4 (Web Text))

Max Activating Token Index: 535

Text #8

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 422022 (C4 (Web Text))

Max Activating Token Index: 943

Text #9

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9575. Min Act: -0.1700

Data Index: 939360 (C4 (Web Text))

Max Activating Token Index: 257

Text #10

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 241419 (C4 (Web Text))

Max Activating Token Index: 623

Text #11

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 832361 (C4 (Web Text))

Max Activating Token Index: 245

Text #12

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0513. Min Act: -0.1700

Data Index: 92480 (C4 (Web Text))

Max Activating Token Index: 140

Text #13

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 87149 (C4 (Web Text))

Max Activating Token Index: 125

Text #14

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0200. Min Act: -0.1700

Data Index: 967495 (C4 (Web Text))

Max Activating Token Index: 253

Text #15

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9263. Min Act: -0.1700

Data Index: 788542 (C4 (Web Text))

Max Activating Token Index: 710

Text #16

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 1237712 (C4 (Web Text))

Max Activating Token Index: 930

Text #17

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 77658 (C4 (Web Text))

Max Activating Token Index: 170

Text #18

Max Range: 5.6763. Min Range: -5.6763

Max Act: 5.0200. Min Act: -0.1700

Data Index: 970155 (C4 (Web Text))

Max Activating Token Index: 767

Text #19

Max Range: 5.6763. Min Range: -5.6763

Max Act: 4.9888. Min Act: -0.1700

Data Index: 819768 (C4 (Web Text))

Max Activating Token Index: 852