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

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

Neuron 509 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: 8.9293. Min Range: -8.9293

Max Act: 8.8668. Min Act: -0.1700

Data Index: 443316 (C4 (Web Text))

Max Activating Token Index: 350

Text #1

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.9293. Min Act: -0.1700

Data Index: 500420 (C4 (Web Text))

Max Activating Token Index: 1004

Text #2

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.7418. Min Act: -0.1700

Data Index: 1040583 (C4 (Web Text))

Max Activating Token Index: 104

Text #3

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.8043. Min Act: -0.1700

Data Index: 163964 (C4 (Web Text))

Max Activating Token Index: 63

Text #4

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.8043. Min Act: -0.1700

Data Index: 547285 (C4 (Web Text))

Max Activating Token Index: 723

Text #5

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.8668. Min Act: -0.1700

Data Index: 1178395 (C4 (Web Text))

Max Activating Token Index: 336

Text #6

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.5543. Min Act: -0.1700

Data Index: 1217122 (C4 (Web Text))

Max Activating Token Index: 775

Text #7

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.6168. Min Act: -0.1700

Data Index: 635270 (C4 (Web Text))

Max Activating Token Index: 965

Text #8

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.4918. Min Act: -0.1700

Data Index: 140489 (C4 (Web Text))

Max Activating Token Index: 982

Text #9

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.6168. Min Act: -0.1700

Data Index: 883501 (C4 (Web Text))

Max Activating Token Index: 601

Text #10

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.5543. Min Act: -0.1700

Data Index: 284897 (C4 (Web Text))

Max Activating Token Index: 734

Text #11

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.4293. Min Act: -0.1700

Data Index: 696623 (C4 (Web Text))

Max Activating Token Index: 442

Text #12

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.4918. Min Act: -0.1700

Data Index: 1199055 (C4 (Web Text))

Max Activating Token Index: 489

Text #13

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.4918. Min Act: -0.1700

Data Index: 74927 (C4 (Web Text))

Max Activating Token Index: 351

Text #14

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.5543. Min Act: -0.1700

Data Index: 349459 (C4 (Web Text))

Max Activating Token Index: 493

Text #15

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.3043. Min Act: -0.1700

Data Index: 1108893 (C4 (Web Text))

Max Activating Token Index: 572

Text #16

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.3668. Min Act: -0.1700

Data Index: 517062 (C4 (Web Text))

Max Activating Token Index: 890

Text #17

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.4918. Min Act: -0.1700

Data Index: 762454 (C4 (Web Text))

Max Activating Token Index: 418

Text #18

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.3668. Min Act: -0.1700

Data Index: 1067811 (C4 (Web Text))

Max Activating Token Index: 154

Text #19

Max Range: 8.9293. Min Range: -8.9293

Max Act: 8.3668. Min Act: -0.1700

Data Index: 725592 (C4 (Web Text))

Max Activating Token Index: 865