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

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

Neuron 132 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: 5.4804. Min Range: -5.4804

Max Act: 5.4804. Min Act: -0.1700

Data Index: 461176 (C4 (Web Text))

Max Activating Token Index: 701

Text #1

Max Range: 5.4804. Min Range: -5.4804

Max Act: 5.1679. Min Act: -0.1700

Data Index: 1251667 (C4 (Web Text))

Max Activating Token Index: 645

Text #2

Max Range: 5.4804. Min Range: -5.4804

Max Act: 5.1679. Min Act: -0.1700

Data Index: 1305625 (C4 (Web Text))

Max Activating Token Index: 829

Text #3

Max Range: 5.4804. Min Range: -5.4804

Max Act: 5.0742. Min Act: -0.1700

Data Index: 236362 (C4 (Web Text))

Max Activating Token Index: 577

Text #4

Max Range: 5.4804. Min Range: -5.4804

Max Act: 5.0117. Min Act: -0.1700

Data Index: 472993 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 5.4804. Min Range: -5.4804

Max Act: 5.0117. Min Act: -0.1700

Data Index: 323861 (C4 (Web Text))

Max Activating Token Index: 649

Text #6

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9492. Min Act: -0.1700

Data Index: 1083329 (C4 (Web Text))

Max Activating Token Index: 558

Text #7

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9804. Min Act: -0.1700

Data Index: 97678 (C4 (Web Text))

Max Activating Token Index: 131

Text #8

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9804. Min Act: -0.1700

Data Index: 1192784 (C4 (Web Text))

Max Activating Token Index: 276

Text #9

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9179. Min Act: -0.1700

Data Index: 844149 (C4 (Web Text))

Max Activating Token Index: 871

Text #10

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8554. Min Act: -0.1700

Data Index: 921262 (C4 (Web Text))

Max Activating Token Index: 670

Text #11

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8554. Min Act: -0.1699

Data Index: 1307708 (C4 (Web Text))

Max Activating Token Index: 318

Text #12

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9492. Min Act: -0.1700

Data Index: 345613 (C4 (Web Text))

Max Activating Token Index: 16

Text #13

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.9179. Min Act: -0.1699

Data Index: 864398 (C4 (Web Text))

Max Activating Token Index: 218

Text #14

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8242. Min Act: -0.1700

Data Index: 47184 (C4 (Web Text))

Max Activating Token Index: 411

Text #15

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8554. Min Act: -0.1700

Data Index: 364 (C4 (Web Text))

Max Activating Token Index: 981

Text #16

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8242. Min Act: -0.1699

Data Index: 1181973 (C4 (Web Text))

Max Activating Token Index: 575

Text #17

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.7929. Min Act: -0.1700

Data Index: 627614 (C4 (Web Text))

Max Activating Token Index: 201

Text #18

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8242. Min Act: -0.1700

Data Index: 79841 (C4 (Web Text))

Max Activating Token Index: 922

Text #19

Max Range: 5.4804. Min Range: -5.4804

Max Act: 4.8242. Min Act: -0.1700

Data Index: 322242 (C4 (Web Text))

Max Activating Token Index: 297