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

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

Neuron 826 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.4789. Min Act: -0.1700

Data Index: 767656 (C4 (Web Text))

Max Activating Token Index: 30

Text #1

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.2289. Min Act: -0.1700

Data Index: 942273 (C4 (Web Text))

Max Activating Token Index: 736

Text #2

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.2289. Min Act: -0.1700

Data Index: 452394 (C4 (Web Text))

Max Activating Token Index: 663

Text #3

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.2289. Min Act: -0.1700

Data Index: 427598 (C4 (Web Text))

Max Activating Token Index: 783

Text #4

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.1664. Min Act: -0.1700

Data Index: 1156130 (C4 (Web Text))

Max Activating Token Index: 748

Text #5

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.1039. Min Act: -0.1700

Data Index: 535981 (C4 (Web Text))

Max Activating Token Index: 240

Text #6

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.1039. Min Act: -0.1700

Data Index: 700254 (C4 (Web Text))

Max Activating Token Index: 342

Text #7

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1700

Data Index: 1048786 (C4 (Web Text))

Max Activating Token Index: 829

Text #8

Max Range: 8.4789. Min Range: -8.4789

Max Act: 8.0414. Min Act: -0.1700

Data Index: 1000457 (C4 (Web Text))

Max Activating Token Index: 66

Text #9

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1700

Data Index: 989861 (C4 (Web Text))

Max Activating Token Index: 211

Text #10

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1677

Data Index: 1115696 (C4 (Web Text))

Max Activating Token Index: 112

Text #11

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1700

Data Index: 653081 (C4 (Web Text))

Max Activating Token Index: 330

Text #12

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1700

Data Index: 806055 (C4 (Web Text))

Max Activating Token Index: 131

Text #13

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1697

Data Index: 394550 (C4 (Web Text))

Max Activating Token Index: 158

Text #14

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9164. Min Act: -0.1700

Data Index: 715442 (C4 (Web Text))

Max Activating Token Index: 491

Text #15

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9164. Min Act: -0.1700

Data Index: 1013461 (C4 (Web Text))

Max Activating Token Index: 248

Text #16

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9164. Min Act: -0.1700

Data Index: 1033014 (C4 (Web Text))

Max Activating Token Index: 449

Text #17

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9164. Min Act: -0.1700

Data Index: 957897 (C4 (Web Text))

Max Activating Token Index: 67

Text #18

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9789. Min Act: -0.1700

Data Index: 676760 (C4 (Web Text))

Max Activating Token Index: 831

Text #19

Max Range: 8.4789. Min Range: -8.4789

Max Act: 7.9164. Min Act: -0.1699

Data Index: 550884 (C4 (Web Text))

Max Activating Token Index: 201