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

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

Neuron 1974 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: 5.8078. Min Range: -5.8078

Max Act: 5.7453. Min Act: -0.1700

Data Index: 446639 (C4 (Web Text))

Max Activating Token Index: 163

Text #1

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.8078. Min Act: -0.1700

Data Index: 1085190 (C4 (Web Text))

Max Activating Token Index: 572

Text #2

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.6516. Min Act: -0.1699

Data Index: 739922 (C4 (Web Text))

Max Activating Token Index: 75

Text #3

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.6516. Min Act: -0.1700

Data Index: 879738 (C4 (Web Text))

Max Activating Token Index: 601

Text #4

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.6516. Min Act: -0.1700

Data Index: 1426187 (Python Code)

Max Activating Token Index: 474

Text #5

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.5578. Min Act: -0.1700

Data Index: 764041 (C4 (Web Text))

Max Activating Token Index: 450

Text #6

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4641. Min Act: -0.1699

Data Index: 405120 (C4 (Web Text))

Max Activating Token Index: 657

Text #7

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.5578. Min Act: -0.1700

Data Index: 1646961 (Python Code)

Max Activating Token Index: 827

Text #8

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.5266. Min Act: -0.1700

Data Index: 778081 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.5266. Min Act: -0.1700

Data Index: 1068760 (C4 (Web Text))

Max Activating Token Index: 190

Text #10

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4953. Min Act: -0.1700

Data Index: 1091170 (C4 (Web Text))

Max Activating Token Index: 209

Text #11

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.5266. Min Act: -0.1700

Data Index: 1470995 (Python Code)

Max Activating Token Index: 465

Text #12

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4953. Min Act: -0.1700

Data Index: 1272642 (C4 (Web Text))

Max Activating Token Index: 584

Text #13

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4641. Min Act: -0.1700

Data Index: 795874 (C4 (Web Text))

Max Activating Token Index: 883

Text #14

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4641. Min Act: -0.1700

Data Index: 217846 (C4 (Web Text))

Max Activating Token Index: 542

Text #15

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4641. Min Act: -0.1699

Data Index: 535995 (C4 (Web Text))

Max Activating Token Index: 206

Text #16

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4953. Min Act: -0.1699

Data Index: 797332 (C4 (Web Text))

Max Activating Token Index: 108

Text #17

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4016. Min Act: -0.1699

Data Index: 483722 (C4 (Web Text))

Max Activating Token Index: 188

Text #18

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.4016. Min Act: -0.1700

Data Index: 816699 (C4 (Web Text))

Max Activating Token Index: 710

Text #19

Max Range: 5.8078. Min Range: -5.8078

Max Act: 5.3078. Min Act: -0.1700

Data Index: 521861 (C4 (Web Text))

Max Activating Token Index: 264