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

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

Neuron 651 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.4697. Min Act: -0.1700

Data Index: 1166460 (C4 (Web Text))

Max Activating Token Index: 502

Text #1

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.2197. Min Act: -0.1700

Data Index: 1021569 (C4 (Web Text))

Max Activating Token Index: 91

Text #2

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.1259. Min Act: -0.1700

Data Index: 1036023 (C4 (Web Text))

Max Activating Token Index: 497

Text #3

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0947. Min Act: -0.1700

Data Index: 166430 (C4 (Web Text))

Max Activating Token Index: 993

Text #4

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0322. Min Act: -0.1700

Data Index: 16068 (C4 (Web Text))

Max Activating Token Index: 862

Text #5

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9697. Min Act: -0.1700

Data Index: 381037 (C4 (Web Text))

Max Activating Token Index: 453

Text #6

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0009. Min Act: -0.1700

Data Index: 751667 (C4 (Web Text))

Max Activating Token Index: 125

Text #7

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0009. Min Act: -0.1700

Data Index: 1282171 (C4 (Web Text))

Max Activating Token Index: 515

Text #8

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9072. Min Act: -0.1700

Data Index: 27243 (C4 (Web Text))

Max Activating Token Index: 558

Text #9

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0009. Min Act: -0.1700

Data Index: 930811 (C4 (Web Text))

Max Activating Token Index: 326

Text #10

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9072. Min Act: -0.1700

Data Index: 1299552 (C4 (Web Text))

Max Activating Token Index: 64

Text #11

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.8759. Min Act: -0.1700

Data Index: 866372 (C4 (Web Text))

Max Activating Token Index: 1000

Text #12

Max Range: 7.4697. Min Range: -7.4697

Max Act: 7.0947. Min Act: -0.1700

Data Index: 1122486 (C4 (Web Text))

Max Activating Token Index: 871

Text #13

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9072. Min Act: -0.1700

Data Index: 1210853 (C4 (Web Text))

Max Activating Token Index: 173

Text #14

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9072. Min Act: -0.1700

Data Index: 1272666 (C4 (Web Text))

Max Activating Token Index: 422

Text #15

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9072. Min Act: -0.1700

Data Index: 146049 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.8759. Min Act: -0.1700

Data Index: 1016712 (C4 (Web Text))

Max Activating Token Index: 889

Text #17

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.8759. Min Act: -0.1700

Data Index: 1316293 (C4 (Web Text))

Max Activating Token Index: 941

Text #18

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.8759. Min Act: -0.1700

Data Index: 110277 (C4 (Web Text))

Max Activating Token Index: 127

Text #19

Max Range: 7.4697. Min Range: -7.4697

Max Act: 6.9384. Min Act: -0.1700

Data Index: 658177 (C4 (Web Text))

Max Activating Token Index: 150