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

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

Neuron 1597 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: 3.1707. Min Range: -3.1707

Max Act: 3.1707. Min Act: -0.1700

Data Index: 662611 (C4 (Web Text))

Max Activating Token Index: 124

Text #1

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1234. Min Act: -0.1700

Data Index: 948259 (C4 (Web Text))

Max Activating Token Index: 736

Text #2

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1234. Min Act: -0.1700

Data Index: 229093 (C4 (Web Text))

Max Activating Token Index: 57

Text #3

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1234. Min Act: -0.1700

Data Index: 579034 (C4 (Web Text))

Max Activating Token Index: 45

Text #4

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1076. Min Act: -0.1700

Data Index: 941582 (C4 (Web Text))

Max Activating Token Index: 145

Text #5

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1392. Min Act: -0.1700

Data Index: 450677 (C4 (Web Text))

Max Activating Token Index: 849

Text #6

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0761. Min Act: -0.1700

Data Index: 551581 (C4 (Web Text))

Max Activating Token Index: 459

Text #7

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.1076. Min Act: -0.1700

Data Index: 922094 (C4 (Web Text))

Max Activating Token Index: 556

Text #8

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0603. Min Act: -0.1700

Data Index: 1487700 (Python Code)

Max Activating Token Index: 45

Text #9

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0603. Min Act: -0.1700

Data Index: 1655062 (Python Code)

Max Activating Token Index: 986

Text #10

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0919. Min Act: -0.1700

Data Index: 670679 (C4 (Web Text))

Max Activating Token Index: 284

Text #11

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0445. Min Act: -0.1700

Data Index: 1240237 (C4 (Web Text))

Max Activating Token Index: 291

Text #12

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0288. Min Act: -0.1700

Data Index: 1210023 (C4 (Web Text))

Max Activating Token Index: 84

Text #13

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0288. Min Act: -0.1700

Data Index: 1062193 (C4 (Web Text))

Max Activating Token Index: 855

Text #14

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0445. Min Act: -0.1700

Data Index: 889069 (C4 (Web Text))

Max Activating Token Index: 681

Text #15

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0761. Min Act: -0.1700

Data Index: 1045957 (C4 (Web Text))

Max Activating Token Index: 94

Text #16

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0603. Min Act: -0.1700

Data Index: 1385306 (Python Code)

Max Activating Token Index: 63

Text #17

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0445. Min Act: -0.1700

Data Index: 249291 (C4 (Web Text))

Max Activating Token Index: 353

Text #18

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0288. Min Act: -0.1700

Data Index: 606286 (C4 (Web Text))

Max Activating Token Index: 224

Text #19

Max Range: 3.1707. Min Range: -3.1707

Max Act: 3.0445. Min Act: -0.1700

Data Index: 1472813 (Python Code)

Max Activating Token Index: 1023