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

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

Neuron 204 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.1128. Min Act: -0.1700

Data Index: 815555 (C4 (Web Text))

Max Activating Token Index: 721

Text #1

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0497. Min Act: -0.1700

Data Index: 1216317 (C4 (Web Text))

Max Activating Token Index: 5

Text #2

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0654. Min Act: -0.1700

Data Index: 645320 (C4 (Web Text))

Max Activating Token Index: 853

Text #3

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0339. Min Act: -0.1700

Data Index: 150075 (C4 (Web Text))

Max Activating Token Index: 996

Text #4

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0181. Min Act: -0.1700

Data Index: 1149010 (C4 (Web Text))

Max Activating Token Index: 564

Text #5

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0181. Min Act: -0.1700

Data Index: 612332 (C4 (Web Text))

Max Activating Token Index: 818

Text #6

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0023. Min Act: -0.1700

Data Index: 1355519 (C4 (Web Text))

Max Activating Token Index: 1001

Text #7

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0023. Min Act: -0.1700

Data Index: 86645 (C4 (Web Text))

Max Activating Token Index: 85

Text #8

Max Range: 3.1128. Min Range: -3.1128

Max Act: 3.0023. Min Act: -0.1700

Data Index: 522932 (C4 (Web Text))

Max Activating Token Index: 761

Text #9

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.9864. Min Act: -0.1700

Data Index: 20280 (C4 (Web Text))

Max Activating Token Index: 117

Text #10

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.9706. Min Act: -0.1700

Data Index: 1018660 (C4 (Web Text))

Max Activating Token Index: 740

Text #11

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.9231. Min Act: -0.1700

Data Index: 48716 (C4 (Web Text))

Max Activating Token Index: 288

Text #12

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.9390. Min Act: -0.1700

Data Index: 676960 (C4 (Web Text))

Max Activating Token Index: 685

Text #13

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8914. Min Act: -0.1700

Data Index: 230694 (C4 (Web Text))

Max Activating Token Index: 239

Text #14

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8914. Min Act: -0.1700

Data Index: 935115 (C4 (Web Text))

Max Activating Token Index: 767

Text #15

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.9073. Min Act: -0.1700

Data Index: 388528 (C4 (Web Text))

Max Activating Token Index: 731

Text #16

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8755. Min Act: -0.1700

Data Index: 1434948 (Python Code)

Max Activating Token Index: 240

Text #17

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8755. Min Act: -0.1700

Data Index: 720403 (C4 (Web Text))

Max Activating Token Index: 1017

Text #18

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8755. Min Act: -0.1700

Data Index: 408918 (C4 (Web Text))

Max Activating Token Index: 427

Text #19

Max Range: 3.1128. Min Range: -3.1128

Max Act: 2.8755. Min Act: -0.1700

Data Index: 1075801 (C4 (Web Text))

Max Activating Token Index: 2