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

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

Neuron 995 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.0274. Min Range: -3.0274

Max Act: 3.0274. Min Act: -0.1700

Data Index: 1002831 (C4 (Web Text))

Max Activating Token Index: 669

Text #1

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.9799. Min Act: -0.1700

Data Index: 1545382 (Python Code)

Max Activating Token Index: 3

Text #2

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.9166. Min Act: -0.1700

Data Index: 135653 (C4 (Web Text))

Max Activating Token Index: 980

Text #3

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.9166. Min Act: -0.1700

Data Index: 1647378 (Python Code)

Max Activating Token Index: 834

Text #4

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.9007. Min Act: -0.1700

Data Index: 1458449 (Python Code)

Max Activating Token Index: 992

Text #5

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8690. Min Act: -0.1700

Data Index: 1542701 (Python Code)

Max Activating Token Index: 29

Text #6

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8372. Min Act: -0.1700

Data Index: 1544705 (Python Code)

Max Activating Token Index: 156

Text #7

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8531. Min Act: -0.1700

Data Index: 120374 (C4 (Web Text))

Max Activating Token Index: 965

Text #8

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8372. Min Act: -0.1700

Data Index: 1647177 (Python Code)

Max Activating Token Index: 903

Text #9

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8372. Min Act: -0.1700

Data Index: 1422784 (Python Code)

Max Activating Token Index: 766

Text #10

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7895. Min Act: -0.1700

Data Index: 1479949 (Python Code)

Max Activating Token Index: 860

Text #11

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7895. Min Act: -0.1700

Data Index: 1479617 (Python Code)

Max Activating Token Index: 111

Text #12

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.8054. Min Act: -0.1700

Data Index: 1593256 (Python Code)

Max Activating Token Index: 671

Text #13

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7735. Min Act: -0.1700

Data Index: 644325 (C4 (Web Text))

Max Activating Token Index: 15

Text #14

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7576. Min Act: -0.1700

Data Index: 809521 (C4 (Web Text))

Max Activating Token Index: 519

Text #15

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7416. Min Act: -0.1700

Data Index: 925616 (C4 (Web Text))

Max Activating Token Index: 190

Text #16

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7416. Min Act: -0.1700

Data Index: 217993 (C4 (Web Text))

Max Activating Token Index: 231

Text #17

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7256. Min Act: -0.1700

Data Index: 830539 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7416. Min Act: -0.1700

Data Index: 1121338 (C4 (Web Text))

Max Activating Token Index: 57

Text #19

Max Range: 3.0274. Min Range: -3.0274

Max Act: 2.7256. Min Act: -0.1700

Data Index: 1156952 (C4 (Web Text))

Max Activating Token Index: 339