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

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

Neuron 1265 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2794. Min Act: -0.1700

Data Index: 720027 (C4 (Web Text))

Max Activating Token Index: 405

Text #1

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.3108. Min Act: -0.1700

Data Index: 217187 (C4 (Web Text))

Max Activating Token Index: 865

Text #2

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2636. Min Act: -0.1700

Data Index: 1132018 (C4 (Web Text))

Max Activating Token Index: 713

Text #3

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2636. Min Act: -0.1700

Data Index: 554726 (C4 (Web Text))

Max Activating Token Index: 106

Text #4

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2322. Min Act: -0.1700

Data Index: 102964 (C4 (Web Text))

Max Activating Token Index: 542

Text #5

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2165. Min Act: -0.1700

Data Index: 1272566 (C4 (Web Text))

Max Activating Token Index: 582

Text #6

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2008. Min Act: -0.1700

Data Index: 307811 (C4 (Web Text))

Max Activating Token Index: 45

Text #7

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2479. Min Act: -0.1700

Data Index: 374815 (C4 (Web Text))

Max Activating Token Index: 656

Text #8

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2008. Min Act: -0.1700

Data Index: 860044 (C4 (Web Text))

Max Activating Token Index: 726

Text #9

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2008. Min Act: -0.1700

Data Index: 766117 (C4 (Web Text))

Max Activating Token Index: 176

Text #10

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2008. Min Act: -0.1700

Data Index: 404810 (C4 (Web Text))

Max Activating Token Index: 642

Text #11

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2165. Min Act: -0.1700

Data Index: 1207948 (C4 (Web Text))

Max Activating Token Index: 628

Text #12

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.2008. Min Act: -0.1700

Data Index: 804627 (C4 (Web Text))

Max Activating Token Index: 131

Text #13

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1850. Min Act: -0.1700

Data Index: 561353 (C4 (Web Text))

Max Activating Token Index: 783

Text #14

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1850. Min Act: -0.1700

Data Index: 961883 (C4 (Web Text))

Max Activating Token Index: 948

Text #15

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1535. Min Act: -0.1700

Data Index: 1265581 (C4 (Web Text))

Max Activating Token Index: 129

Text #16

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1063. Min Act: -0.1700

Data Index: 568855 (C4 (Web Text))

Max Activating Token Index: 464

Text #17

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1693. Min Act: -0.1700

Data Index: 1207498 (C4 (Web Text))

Max Activating Token Index: 792

Text #18

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1378. Min Act: -0.1700

Data Index: 195395 (C4 (Web Text))

Max Activating Token Index: 384

Text #19

Max Range: 3.3108. Min Range: -3.3108

Max Act: 3.1220. Min Act: -0.1700

Data Index: 1605906 (Python Code)

Max Activating Token Index: 956