Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 704 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7218. Min Act: -0.1700

Data Index: 220957 (C4 (Web Text))

Max Activating Token Index: 195

Text #1

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7531. Min Act: -0.1700

Data Index: 865684 (C4 (Web Text))

Max Activating Token Index: 635

Text #2

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7218. Min Act: -0.1700

Data Index: 658074 (C4 (Web Text))

Max Activating Token Index: 195

Text #3

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7218. Min Act: -0.1700

Data Index: 644450 (C4 (Web Text))

Max Activating Token Index: 696

Text #4

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7531. Min Act: -0.1700

Data Index: 696141 (C4 (Web Text))

Max Activating Token Index: 726

Text #5

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7218. Min Act: -0.1700

Data Index: 1148024 (C4 (Web Text))

Max Activating Token Index: 241

Text #6

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6905. Min Act: -0.1700

Data Index: 519315 (C4 (Web Text))

Max Activating Token Index: 612

Text #7

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6905. Min Act: -0.1700

Data Index: 160113 (C4 (Web Text))

Max Activating Token Index: 908

Text #8

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6905. Min Act: -0.1700

Data Index: 568258 (C4 (Web Text))

Max Activating Token Index: 290

Text #9

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6905. Min Act: -0.1700

Data Index: 597768 (C4 (Web Text))

Max Activating Token Index: 57

Text #10

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.7218. Min Act: -0.1700

Data Index: 22525 (C4 (Web Text))

Max Activating Token Index: 114

Text #11

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6749. Min Act: -0.1700

Data Index: 1112887 (C4 (Web Text))

Max Activating Token Index: 890

Text #12

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6749. Min Act: -0.1700

Data Index: 567125 (C4 (Web Text))

Max Activating Token Index: 243

Text #13

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6592. Min Act: -0.1700

Data Index: 874123 (C4 (Web Text))

Max Activating Token Index: 389

Text #14

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6592. Min Act: -0.1700

Data Index: 781413 (C4 (Web Text))

Max Activating Token Index: 947

Text #15

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6592. Min Act: -0.1700

Data Index: 282408 (C4 (Web Text))

Max Activating Token Index: 527

Text #16

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6592. Min Act: -0.1700

Data Index: 148562 (C4 (Web Text))

Max Activating Token Index: 412

Text #17

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6436. Min Act: -0.1700

Data Index: 847168 (C4 (Web Text))

Max Activating Token Index: 182

Text #18

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6279. Min Act: -0.1700

Data Index: 567386 (C4 (Web Text))

Max Activating Token Index: 171

Text #19

Max Range: 3.7531. Min Range: -3.7531

Max Act: 3.6436. Min Act: -0.1700

Data Index: 268706 (C4 (Web Text))

Max Activating Token Index: 341