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

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

Neuron 1311 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: 6.7007. Min Range: -6.7007

Max Act: 6.7007. Min Act: -0.1673

Data Index: 152827 (C4 (Web Text))

Max Activating Token Index: 180

Text #1

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.6069. Min Act: -0.1680

Data Index: 953002 (C4 (Web Text))

Max Activating Token Index: 520

Text #2

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.5757. Min Act: -0.1691

Data Index: 268657 (C4 (Web Text))

Max Activating Token Index: 503

Text #3

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4819. Min Act: -0.1597

Data Index: 273292 (C4 (Web Text))

Max Activating Token Index: 355

Text #4

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4819. Min Act: -0.1700

Data Index: 218640 (C4 (Web Text))

Max Activating Token Index: 374

Text #5

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4507. Min Act: -0.1597

Data Index: 1308133 (C4 (Web Text))

Max Activating Token Index: 391

Text #6

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4507. Min Act: -0.1689

Data Index: 417331 (C4 (Web Text))

Max Activating Token Index: 751

Text #7

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4507. Min Act: -0.1682

Data Index: 935194 (C4 (Web Text))

Max Activating Token Index: 296

Text #8

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.3882. Min Act: -0.1700

Data Index: 615020 (C4 (Web Text))

Max Activating Token Index: 42

Text #9

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.4194. Min Act: -0.1700

Data Index: 1079146 (C4 (Web Text))

Max Activating Token Index: 94

Text #10

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.3569. Min Act: -0.1700

Data Index: 147821 (C4 (Web Text))

Max Activating Token Index: 900

Text #11

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.3257. Min Act: -0.1700

Data Index: 563164 (C4 (Web Text))

Max Activating Token Index: 573

Text #12

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2944. Min Act: -0.1696

Data Index: 578283 (C4 (Web Text))

Max Activating Token Index: 1014

Text #13

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2632. Min Act: -0.1700

Data Index: 113245 (C4 (Web Text))

Max Activating Token Index: 873

Text #14

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.3257. Min Act: -0.1700

Data Index: 425082 (C4 (Web Text))

Max Activating Token Index: 488

Text #15

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2944. Min Act: -0.1700

Data Index: 967295 (C4 (Web Text))

Max Activating Token Index: 333

Text #16

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2632. Min Act: -0.1700

Data Index: 855785 (C4 (Web Text))

Max Activating Token Index: 951

Text #17

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2632. Min Act: -0.1621

Data Index: 1179188 (C4 (Web Text))

Max Activating Token Index: 34

Text #18

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2632. Min Act: -0.1700

Data Index: 230496 (C4 (Web Text))

Max Activating Token Index: 982

Text #19

Max Range: 6.7007. Min Range: -6.7007

Max Act: 6.2319. Min Act: -0.1700

Data Index: 1139361 (C4 (Web Text))

Max Activating Token Index: 27