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

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

Neuron 1981 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: 7.7178. Min Range: -7.7178

Max Act: 7.7178. Min Act: -0.1700

Data Index: 1322254 (C4 (Web Text))

Max Activating Token Index: 75

Text #1

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.6866. Min Act: -0.1700

Data Index: 1181161 (C4 (Web Text))

Max Activating Token Index: 810

Text #2

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.4991. Min Act: -0.1700

Data Index: 1298269 (C4 (Web Text))

Max Activating Token Index: 1011

Text #3

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.4991. Min Act: -0.1700

Data Index: 1161268 (C4 (Web Text))

Max Activating Token Index: 617

Text #4

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.4991. Min Act: -0.1700

Data Index: 1144525 (C4 (Web Text))

Max Activating Token Index: 384

Text #5

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.4991. Min Act: -0.1700

Data Index: 17926 (C4 (Web Text))

Max Activating Token Index: 93

Text #6

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.5303. Min Act: -0.1700

Data Index: 1301546 (C4 (Web Text))

Max Activating Token Index: 467

Text #7

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.4053. Min Act: -0.1700

Data Index: 595680 (C4 (Web Text))

Max Activating Token Index: 133

Text #8

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.3116. Min Act: -0.1700

Data Index: 255368 (C4 (Web Text))

Max Activating Token Index: 970

Text #9

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.2491. Min Act: -0.1700

Data Index: 518469 (C4 (Web Text))

Max Activating Token Index: 430

Text #10

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.3428. Min Act: -0.1700

Data Index: 199180 (C4 (Web Text))

Max Activating Token Index: 597

Text #11

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.2803. Min Act: -0.1700

Data Index: 909248 (C4 (Web Text))

Max Activating Token Index: 722

Text #12

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.2803. Min Act: -0.1700

Data Index: 145277 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.2491. Min Act: -0.1700

Data Index: 952116 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.1553. Min Act: -0.1700

Data Index: 179119 (C4 (Web Text))

Max Activating Token Index: 556

Text #15

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.1241. Min Act: -0.1700

Data Index: 479341 (C4 (Web Text))

Max Activating Token Index: 665

Text #16

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.0928. Min Act: -0.1700

Data Index: 1020203 (C4 (Web Text))

Max Activating Token Index: 1015

Text #17

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.1553. Min Act: -0.1700

Data Index: 958324 (C4 (Web Text))

Max Activating Token Index: 517

Text #18

Max Range: 7.7178. Min Range: -7.7178

Max Act: 7.1241. Min Act: -0.1700

Data Index: 213377 (C4 (Web Text))

Max Activating Token Index: 493

Text #19

Max Range: 7.7178. Min Range: -7.7178

Max Act: 6.9991. Min Act: -0.1700

Data Index: 607065 (C4 (Web Text))

Max Activating Token Index: 411