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

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

Neuron 767 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: 7.1677. Min Range: -7.1677

Max Act: 7.1677. Min Act: -0.1700

Data Index: 1183530 (C4 (Web Text))

Max Activating Token Index: 215

Text #1

Max Range: 7.1677. Min Range: -7.1677

Max Act: 7.1677. Min Act: -0.1700

Data Index: 1200724 (C4 (Web Text))

Max Activating Token Index: 63

Text #2

Max Range: 7.1677. Min Range: -7.1677

Max Act: 7.0114. Min Act: -0.1700

Data Index: 471406 (C4 (Web Text))

Max Activating Token Index: 804

Text #3

Max Range: 7.1677. Min Range: -7.1677

Max Act: 7.0427. Min Act: -0.1700

Data Index: 569211 (C4 (Web Text))

Max Activating Token Index: 393

Text #4

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.9802. Min Act: -0.1700

Data Index: 364059 (C4 (Web Text))

Max Activating Token Index: 343

Text #5

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.6989. Min Act: -0.1700

Data Index: 775865 (C4 (Web Text))

Max Activating Token Index: 207

Text #6

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.6989. Min Act: -0.1695

Data Index: 119830 (C4 (Web Text))

Max Activating Token Index: 894

Text #7

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.5427. Min Act: -0.1700

Data Index: 1141480 (C4 (Web Text))

Max Activating Token Index: 256

Text #8

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.5427. Min Act: -0.1700

Data Index: 971049 (C4 (Web Text))

Max Activating Token Index: 594

Text #9

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.5427. Min Act: -0.1700

Data Index: 32170 (C4 (Web Text))

Max Activating Token Index: 866

Text #10

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.5739. Min Act: -0.1700

Data Index: 442455 (C4 (Web Text))

Max Activating Token Index: 625

Text #11

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4802. Min Act: -0.1700

Data Index: 220473 (C4 (Web Text))

Max Activating Token Index: 493

Text #12

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4802. Min Act: -0.1700

Data Index: 929421 (C4 (Web Text))

Max Activating Token Index: 241

Text #13

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4489. Min Act: -0.1700

Data Index: 1105134 (C4 (Web Text))

Max Activating Token Index: 114

Text #14

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4489. Min Act: -0.1700

Data Index: 999703 (C4 (Web Text))

Max Activating Token Index: 388

Text #15

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4802. Min Act: -0.1700

Data Index: 758493 (C4 (Web Text))

Max Activating Token Index: 254

Text #16

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4802. Min Act: -0.1700

Data Index: 919153 (C4 (Web Text))

Max Activating Token Index: 123

Text #17

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4802. Min Act: -0.1700

Data Index: 480634 (C4 (Web Text))

Max Activating Token Index: 122

Text #18

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.4177. Min Act: -0.1700

Data Index: 1012921 (C4 (Web Text))

Max Activating Token Index: 25

Text #19

Max Range: 7.1677. Min Range: -7.1677

Max Act: 6.3864. Min Act: -0.1700

Data Index: 402169 (C4 (Web Text))

Max Activating Token Index: 762