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

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

Neuron 1538 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: 7.1947. Min Range: -7.1947

Max Act: 7.1947. Min Act: -0.1700

Data Index: 270740 (C4 (Web Text))

Max Activating Token Index: 784

Text #1

Max Range: 7.1947. Min Range: -7.1947

Max Act: 7.1322. Min Act: -0.1700

Data Index: 546273 (C4 (Web Text))

Max Activating Token Index: 205

Text #2

Max Range: 7.1947. Min Range: -7.1947

Max Act: 7.0072. Min Act: -0.1700

Data Index: 804437 (C4 (Web Text))

Max Activating Token Index: 260

Text #3

Max Range: 7.1947. Min Range: -7.1947

Max Act: 7.1322. Min Act: -0.1700

Data Index: 699629 (C4 (Web Text))

Max Activating Token Index: 627

Text #4

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.7572. Min Act: -0.1700

Data Index: 694433 (C4 (Web Text))

Max Activating Token Index: 520

Text #5

Max Range: 7.1947. Min Range: -7.1947

Max Act: 7.0697. Min Act: -0.1700

Data Index: 874963 (C4 (Web Text))

Max Activating Token Index: 167

Text #6

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.7572. Min Act: -0.1700

Data Index: 159015 (C4 (Web Text))

Max Activating Token Index: 792

Text #7

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6947. Min Act: -0.1700

Data Index: 886530 (C4 (Web Text))

Max Activating Token Index: 243

Text #8

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6322. Min Act: -0.1700

Data Index: 798699 (C4 (Web Text))

Max Activating Token Index: 486

Text #9

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.7572. Min Act: -0.1700

Data Index: 1220497 (C4 (Web Text))

Max Activating Token Index: 267

Text #10

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.7572. Min Act: -0.1700

Data Index: 625284 (C4 (Web Text))

Max Activating Token Index: 917

Text #11

Max Range: 7.1947. Min Range: -7.1947

Max Act: 7.0072. Min Act: -0.1700

Data Index: 1123355 (C4 (Web Text))

Max Activating Token Index: 396

Text #12

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6947. Min Act: -0.1700

Data Index: 790674 (C4 (Web Text))

Max Activating Token Index: 305

Text #13

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.8197. Min Act: -0.1700

Data Index: 1054999 (C4 (Web Text))

Max Activating Token Index: 392

Text #14

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.8197. Min Act: -0.1700

Data Index: 1221418 (C4 (Web Text))

Max Activating Token Index: 195

Text #15

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6947. Min Act: -0.1700

Data Index: 371 (C4 (Web Text))

Max Activating Token Index: 523

Text #16

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6947. Min Act: -0.1700

Data Index: 338599 (C4 (Web Text))

Max Activating Token Index: 552

Text #17

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.7572. Min Act: -0.1700

Data Index: 869744 (C4 (Web Text))

Max Activating Token Index: 696

Text #18

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6322. Min Act: -0.1700

Data Index: 575692 (C4 (Web Text))

Max Activating Token Index: 637

Text #19

Max Range: 7.1947. Min Range: -7.1947

Max Act: 6.6009. Min Act: -0.1700

Data Index: 984992 (C4 (Web Text))

Max Activating Token Index: 495