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

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

Neuron 1168 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: 3.5976. Min Range: -3.5976

Max Act: 3.5976. Min Act: -0.1700

Data Index: 1180761 (C4 (Web Text))

Max Activating Token Index: 395

Text #1

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.4253. Min Act: -0.1700

Data Index: 311211 (C4 (Web Text))

Max Activating Token Index: 465

Text #2

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.3468. Min Act: -0.1700

Data Index: 1261950 (C4 (Web Text))

Max Activating Token Index: 851

Text #3

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.3154. Min Act: -0.1700

Data Index: 829947 (C4 (Web Text))

Max Activating Token Index: 914

Text #4

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.3154. Min Act: -0.1700

Data Index: 91593 (C4 (Web Text))

Max Activating Token Index: 156

Text #5

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2840. Min Act: -0.1700

Data Index: 1582911 (Python Code)

Max Activating Token Index: 410

Text #6

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2997. Min Act: -0.1700

Data Index: 788287 (C4 (Web Text))

Max Activating Token Index: 279

Text #7

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2683. Min Act: -0.1700

Data Index: 21956 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2526. Min Act: -0.1700

Data Index: 762013 (C4 (Web Text))

Max Activating Token Index: 405

Text #9

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2526. Min Act: -0.1700

Data Index: 505041 (C4 (Web Text))

Max Activating Token Index: 576

Text #10

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2683. Min Act: -0.1700

Data Index: 1193356 (C4 (Web Text))

Max Activating Token Index: 319

Text #11

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2683. Min Act: -0.1700

Data Index: 320972 (C4 (Web Text))

Max Activating Token Index: 21

Text #12

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2211. Min Act: -0.1700

Data Index: 626619 (C4 (Web Text))

Max Activating Token Index: 287

Text #13

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2369. Min Act: -0.1700

Data Index: 588322 (C4 (Web Text))

Max Activating Token Index: 356

Text #14

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2211. Min Act: -0.1700

Data Index: 1323653 (C4 (Web Text))

Max Activating Token Index: 564

Text #15

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2211. Min Act: -0.1700

Data Index: 96786 (C4 (Web Text))

Max Activating Token Index: 892

Text #16

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2211. Min Act: -0.1700

Data Index: 870785 (C4 (Web Text))

Max Activating Token Index: 978

Text #17

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2054. Min Act: -0.1700

Data Index: 681194 (C4 (Web Text))

Max Activating Token Index: 210

Text #18

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2054. Min Act: -0.1700

Data Index: 499207 (C4 (Web Text))

Max Activating Token Index: 610

Text #19

Max Range: 3.5976. Min Range: -3.5976

Max Act: 3.2054. Min Act: -0.1700

Data Index: 617796 (C4 (Web Text))

Max Activating Token Index: 763