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

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

Neuron 2010 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: 5.4177. Min Range: -5.4177

Max Act: 5.4177. Min Act: -0.1700

Data Index: 1155722 (C4 (Web Text))

Max Activating Token Index: 781

Text #1

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.3240. Min Act: -0.1700

Data Index: 725249 (C4 (Web Text))

Max Activating Token Index: 489

Text #2

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.3552. Min Act: -0.1700

Data Index: 154759 (C4 (Web Text))

Max Activating Token Index: 294

Text #3

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.3240. Min Act: -0.1700

Data Index: 1238330 (C4 (Web Text))

Max Activating Token Index: 95

Text #4

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.3240. Min Act: -0.1700

Data Index: 1320694 (C4 (Web Text))

Max Activating Token Index: 511

Text #5

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.2302. Min Act: -0.1700

Data Index: 1233424 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.2302. Min Act: -0.1700

Data Index: 1061047 (C4 (Web Text))

Max Activating Token Index: 244

Text #7

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.2302. Min Act: -0.1700

Data Index: 125708 (C4 (Web Text))

Max Activating Token Index: 815

Text #8

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 398898 (C4 (Web Text))

Max Activating Token Index: 122

Text #9

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 1161507 (C4 (Web Text))

Max Activating Token Index: 596

Text #10

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 890309 (C4 (Web Text))

Max Activating Token Index: 179

Text #11

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 1020764 (C4 (Web Text))

Max Activating Token Index: 829

Text #12

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 124378 (C4 (Web Text))

Max Activating Token Index: 200

Text #13

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 167820 (C4 (Web Text))

Max Activating Token Index: 121

Text #14

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.2302. Min Act: -0.1700

Data Index: 129627 (C4 (Web Text))

Max Activating Token Index: 374

Text #15

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1677. Min Act: -0.1700

Data Index: 242547 (C4 (Web Text))

Max Activating Token Index: 199

Text #16

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1677. Min Act: -0.1700

Data Index: 670994 (C4 (Web Text))

Max Activating Token Index: 276

Text #17

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1365. Min Act: -0.1700

Data Index: 1007684 (C4 (Web Text))

Max Activating Token Index: 900

Text #18

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 1024977 (C4 (Web Text))

Max Activating Token Index: 163

Text #19

Max Range: 5.4177. Min Range: -5.4177

Max Act: 5.1990. Min Act: -0.1700

Data Index: 928557 (C4 (Web Text))

Max Activating Token Index: 611