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

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

Neuron 320 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: 5.8864. Min Range: -5.8864

Max Act: 5.8864. Min Act: -0.1700

Data Index: 567999 (C4 (Web Text))

Max Activating Token Index: 183

Text #1

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.3552. Min Act: -0.1700

Data Index: 1064674 (C4 (Web Text))

Max Activating Token Index: 697

Text #2

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 241192 (C4 (Web Text))

Max Activating Token Index: 902

Text #3

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1677. Min Act: -0.1700

Data Index: 406694 (C4 (Web Text))

Max Activating Token Index: 688

Text #4

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.2614. Min Act: -0.1700

Data Index: 404337 (C4 (Web Text))

Max Activating Token Index: 946

Text #5

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 1121463 (C4 (Web Text))

Max Activating Token Index: 339

Text #6

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.2302. Min Act: -0.1700

Data Index: 1129007 (C4 (Web Text))

Max Activating Token Index: 160

Text #7

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 371192 (C4 (Web Text))

Max Activating Token Index: 911

Text #8

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 1320772 (C4 (Web Text))

Max Activating Token Index: 356

Text #9

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0739. Min Act: -0.1700

Data Index: 612275 (C4 (Web Text))

Max Activating Token Index: 769

Text #10

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0427. Min Act: -0.1700

Data Index: 1049043 (C4 (Web Text))

Max Activating Token Index: 736

Text #11

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0739. Min Act: -0.1700

Data Index: 482576 (C4 (Web Text))

Max Activating Token Index: 592

Text #12

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0427. Min Act: -0.1700

Data Index: 1144951 (C4 (Web Text))

Max Activating Token Index: 155

Text #13

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 1136923 (C4 (Web Text))

Max Activating Token Index: 431

Text #14

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0427. Min Act: -0.1700

Data Index: 290519 (C4 (Web Text))

Max Activating Token Index: 54

Text #15

Max Range: 5.8864. Min Range: -5.8864

Max Act: 4.9802. Min Act: -0.1700

Data Index: 735936 (C4 (Web Text))

Max Activating Token Index: 930

Text #16

Max Range: 5.8864. Min Range: -5.8864

Max Act: 4.9489. Min Act: -0.1700

Data Index: 225709 (C4 (Web Text))

Max Activating Token Index: 218

Text #17

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.1364. Min Act: -0.1700

Data Index: 193005 (C4 (Web Text))

Max Activating Token Index: 369

Text #18

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0427. Min Act: -0.1700

Data Index: 485799 (C4 (Web Text))

Max Activating Token Index: 576

Text #19

Max Range: 5.8864. Min Range: -5.8864

Max Act: 5.0114. Min Act: -0.1700

Data Index: 1182103 (C4 (Web Text))

Max Activating Token Index: 435