Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1340 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 7.0270. Min Range: -7.0270

Max Act: 7.0270. Min Act: -0.1700

Data Index: 1308151 (C4 (Web Text))

Max Activating Token Index: 382

Text #1

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.7145. Min Act: -0.1700

Data Index: 767948 (C4 (Web Text))

Max Activating Token Index: 813

Text #2

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.5583. Min Act: -0.1700

Data Index: 520658 (C4 (Web Text))

Max Activating Token Index: 645

Text #3

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.4958. Min Act: -0.1700

Data Index: 1185865 (C4 (Web Text))

Max Activating Token Index: 720

Text #4

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.4645. Min Act: -0.1700

Data Index: 427813 (C4 (Web Text))

Max Activating Token Index: 984

Text #5

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.5895. Min Act: -0.1700

Data Index: 331532 (C4 (Web Text))

Max Activating Token Index: 273

Text #6

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.3395. Min Act: -0.1700

Data Index: 553166 (C4 (Web Text))

Max Activating Token Index: 800

Text #7

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.4020. Min Act: -0.1700

Data Index: 909919 (C4 (Web Text))

Max Activating Token Index: 182

Text #8

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.4333. Min Act: -0.1700

Data Index: 141436 (C4 (Web Text))

Max Activating Token Index: 696

Text #9

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.3708. Min Act: -0.1700

Data Index: 1183916 (C4 (Web Text))

Max Activating Token Index: 839

Text #10

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.2770. Min Act: -0.1700

Data Index: 1129847 (C4 (Web Text))

Max Activating Token Index: 15

Text #11

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.3395. Min Act: -0.1700

Data Index: 763215 (C4 (Web Text))

Max Activating Token Index: 41

Text #12

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.3395. Min Act: -0.1700

Data Index: 286304 (C4 (Web Text))

Max Activating Token Index: 514

Text #13

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.3395. Min Act: -0.1700

Data Index: 1268630 (C4 (Web Text))

Max Activating Token Index: 906

Text #14

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.2458. Min Act: -0.1700

Data Index: 1307137 (C4 (Web Text))

Max Activating Token Index: 178

Text #15

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.1833. Min Act: -0.1700

Data Index: 1474796 (Python Code)

Max Activating Token Index: 307

Text #16

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.1833. Min Act: -0.1700

Data Index: 112971 (C4 (Web Text))

Max Activating Token Index: 952

Text #17

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.1833. Min Act: -0.1700

Data Index: 904304 (C4 (Web Text))

Max Activating Token Index: 685

Text #18

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.1520. Min Act: -0.1700

Data Index: 870601 (C4 (Web Text))

Max Activating Token Index: 981

Text #19

Max Range: 7.0270. Min Range: -7.0270

Max Act: 6.2145. Min Act: -0.1700

Data Index: 965241 (C4 (Web Text))

Max Activating Token Index: 364