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

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

Neuron 1628 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: 3.7145. Min Range: -3.7145

Max Act: 3.7145. Min Act: -0.1700

Data Index: 716981 (C4 (Web Text))

Max Activating Token Index: 465

Text #1

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6832. Min Act: -0.1700

Data Index: 647671 (C4 (Web Text))

Max Activating Token Index: 237

Text #2

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6519. Min Act: -0.1700

Data Index: 282544 (C4 (Web Text))

Max Activating Token Index: 528

Text #3

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6519. Min Act: -0.1700

Data Index: 43747 (C4 (Web Text))

Max Activating Token Index: 251

Text #4

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6519. Min Act: -0.1700

Data Index: 52148 (C4 (Web Text))

Max Activating Token Index: 869

Text #5

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6206. Min Act: -0.1700

Data Index: 476075 (C4 (Web Text))

Max Activating Token Index: 468

Text #6

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.6206. Min Act: -0.1700

Data Index: 656614 (C4 (Web Text))

Max Activating Token Index: 945

Text #7

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5893. Min Act: -0.1700

Data Index: 1245695 (C4 (Web Text))

Max Activating Token Index: 275

Text #8

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5893. Min Act: -0.1700

Data Index: 4194 (C4 (Web Text))

Max Activating Token Index: 943

Text #9

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5580. Min Act: -0.1700

Data Index: 1168496 (C4 (Web Text))

Max Activating Token Index: 634

Text #10

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5266. Min Act: -0.1700

Data Index: 148768 (C4 (Web Text))

Max Activating Token Index: 820

Text #11

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5580. Min Act: -0.1700

Data Index: 647547 (C4 (Web Text))

Max Activating Token Index: 824

Text #12

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5580. Min Act: -0.1700

Data Index: 34781 (C4 (Web Text))

Max Activating Token Index: 203

Text #13

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5580. Min Act: -0.1700

Data Index: 681511 (C4 (Web Text))

Max Activating Token Index: 72

Text #14

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5266. Min Act: -0.1700

Data Index: 276084 (C4 (Web Text))

Max Activating Token Index: 635

Text #15

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5893. Min Act: -0.1700

Data Index: 1519982 (Python Code)

Max Activating Token Index: 886

Text #16

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5266. Min Act: -0.1700

Data Index: 558747 (C4 (Web Text))

Max Activating Token Index: 430

Text #17

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.4953. Min Act: -0.1700

Data Index: 435260 (C4 (Web Text))

Max Activating Token Index: 141

Text #18

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5266. Min Act: -0.1700

Data Index: 441082 (C4 (Web Text))

Max Activating Token Index: 115

Text #19

Max Range: 3.7145. Min Range: -3.7145

Max Act: 3.5266. Min Act: -0.1700

Data Index: 42981 (C4 (Web Text))

Max Activating Token Index: 98