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

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

Neuron 90 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.8137. Min Act: -0.1700

Data Index: 1565761 (Python Code)

Max Activating Token Index: 261

Text #1

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.3595. Min Act: -0.1700

Data Index: 379604 (C4 (Web Text))

Max Activating Token Index: 593

Text #2

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.3438. Min Act: -0.1700

Data Index: 1175018 (C4 (Web Text))

Max Activating Token Index: 879

Text #3

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.3438. Min Act: -0.1700

Data Index: 167651 (C4 (Web Text))

Max Activating Token Index: 563

Text #4

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2967. Min Act: -0.1700

Data Index: 1264692 (C4 (Web Text))

Max Activating Token Index: 601

Text #5

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.3124. Min Act: -0.1700

Data Index: 365965 (C4 (Web Text))

Max Activating Token Index: 489

Text #6

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2653. Min Act: -0.1700

Data Index: 1051229 (C4 (Web Text))

Max Activating Token Index: 496

Text #7

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2338. Min Act: -0.1700

Data Index: 925639 (C4 (Web Text))

Max Activating Token Index: 102

Text #8

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2496. Min Act: -0.1700

Data Index: 277794 (C4 (Web Text))

Max Activating Token Index: 530

Text #9

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.1866. Min Act: -0.1700

Data Index: 745528 (C4 (Web Text))

Max Activating Token Index: 586

Text #10

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2181. Min Act: -0.1700

Data Index: 984513 (C4 (Web Text))

Max Activating Token Index: 407

Text #11

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2338. Min Act: -0.1700

Data Index: 582055 (C4 (Web Text))

Max Activating Token Index: 474

Text #12

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2181. Min Act: -0.1700

Data Index: 166706 (C4 (Web Text))

Max Activating Token Index: 971

Text #13

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2181. Min Act: -0.1700

Data Index: 816981 (C4 (Web Text))

Max Activating Token Index: 376

Text #14

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2181. Min Act: -0.1700

Data Index: 345847 (C4 (Web Text))

Max Activating Token Index: 132

Text #15

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2338. Min Act: -0.1700

Data Index: 216137 (C4 (Web Text))

Max Activating Token Index: 292

Text #16

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2024. Min Act: -0.1700

Data Index: 626522 (C4 (Web Text))

Max Activating Token Index: 35

Text #17

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.1866. Min Act: -0.1700

Data Index: 734603 (C4 (Web Text))

Max Activating Token Index: 412

Text #18

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2181. Min Act: -0.1700

Data Index: 977826 (C4 (Web Text))

Max Activating Token Index: 996

Text #19

Max Range: 3.8137. Min Range: -3.8137

Max Act: 3.2024. Min Act: -0.1700

Data Index: 1292569 (C4 (Web Text))

Max Activating Token Index: 55