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

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

Neuron 1735 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: 7.8310. Min Range: -7.8310

Max Act: 7.8310. Min Act: -0.1700

Data Index: 24910 (C4 (Web Text))

Max Activating Token Index: 554

Text #1

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.7997. Min Act: -0.1700

Data Index: 1276805 (C4 (Web Text))

Max Activating Token Index: 364

Text #2

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.5810. Min Act: -0.1700

Data Index: 819287 (C4 (Web Text))

Max Activating Token Index: 238

Text #3

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.5497. Min Act: -0.1700

Data Index: 795090 (C4 (Web Text))

Max Activating Token Index: 303

Text #4

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.4872. Min Act: -0.1700

Data Index: 38686 (C4 (Web Text))

Max Activating Token Index: 187

Text #5

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.5185. Min Act: -0.1700

Data Index: 883963 (C4 (Web Text))

Max Activating Token Index: 472

Text #6

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.5185. Min Act: -0.1700

Data Index: 491626 (C4 (Web Text))

Max Activating Token Index: 96

Text #7

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.3935. Min Act: -0.1700

Data Index: 1072736 (C4 (Web Text))

Max Activating Token Index: 482

Text #8

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.3935. Min Act: -0.1700

Data Index: 1075897 (C4 (Web Text))

Max Activating Token Index: 379

Text #9

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.3622. Min Act: -0.1700

Data Index: 234843 (C4 (Web Text))

Max Activating Token Index: 66

Text #10

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9872. Min Act: -0.1700

Data Index: 1374965 (Python Code)

Max Activating Token Index: 688

Text #11

Max Range: 7.8310. Min Range: -7.8310

Max Act: 7.0185. Min Act: -0.1700

Data Index: 1220857 (C4 (Web Text))

Max Activating Token Index: 1020

Text #12

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9560. Min Act: -0.1700

Data Index: 382079 (C4 (Web Text))

Max Activating Token Index: 648

Text #13

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9560. Min Act: -0.1699

Data Index: 289128 (C4 (Web Text))

Max Activating Token Index: 332

Text #14

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9872. Min Act: -0.1700

Data Index: 1459045 (Python Code)

Max Activating Token Index: 110

Text #15

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9247. Min Act: -0.1699

Data Index: 1454102 (Python Code)

Max Activating Token Index: 153

Text #16

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.7685. Min Act: -0.1700

Data Index: 332644 (C4 (Web Text))

Max Activating Token Index: 793

Text #17

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.9872. Min Act: -0.1700

Data Index: 1453057 (Python Code)

Max Activating Token Index: 173

Text #18

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.8310. Min Act: -0.1700

Data Index: 1650454 (Python Code)

Max Activating Token Index: 176

Text #19

Max Range: 7.8310. Min Range: -7.8310

Max Act: 6.7060. Min Act: -0.1700

Data Index: 1160192 (C4 (Web Text))

Max Activating Token Index: 1001