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

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

Neuron 1813 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: 5.4355. Min Range: -5.4355

Max Act: 5.4355. Min Act: -0.1700

Data Index: 546329 (C4 (Web Text))

Max Activating Token Index: 841

Text #1

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.3105. Min Act: -0.1700

Data Index: 784341 (C4 (Web Text))

Max Activating Token Index: 371

Text #2

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.2480. Min Act: -0.1700

Data Index: 377672 (C4 (Web Text))

Max Activating Token Index: 183

Text #3

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.2480. Min Act: -0.1700

Data Index: 1268623 (C4 (Web Text))

Max Activating Token Index: 175

Text #4

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.2480. Min Act: -0.1700

Data Index: 1285564 (C4 (Web Text))

Max Activating Token Index: 263

Text #5

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.2167. Min Act: -0.1700

Data Index: 157511 (C4 (Web Text))

Max Activating Token Index: 252

Text #6

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.1230. Min Act: -0.1700

Data Index: 161417 (C4 (Web Text))

Max Activating Token Index: 515

Text #7

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.1230. Min Act: -0.1700

Data Index: 918608 (C4 (Web Text))

Max Activating Token Index: 826

Text #8

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.1230. Min Act: -0.1700

Data Index: 260641 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.1230. Min Act: -0.1700

Data Index: 133880 (C4 (Web Text))

Max Activating Token Index: 747

Text #10

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.0605. Min Act: -0.1700

Data Index: 659310 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.0917. Min Act: -0.1700

Data Index: 1006410 (C4 (Web Text))

Max Activating Token Index: 482

Text #12

Max Range: 5.4355. Min Range: -5.4355

Max Act: 4.9980. Min Act: -0.1700

Data Index: 110309 (C4 (Web Text))

Max Activating Token Index: 796

Text #13

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.0605. Min Act: -0.1700

Data Index: 428087 (C4 (Web Text))

Max Activating Token Index: 903

Text #14

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.0292. Min Act: -0.1700

Data Index: 1416881 (Python Code)

Max Activating Token Index: 12

Text #15

Max Range: 5.4355. Min Range: -5.4355

Max Act: 4.9980. Min Act: -0.1700

Data Index: 137126 (C4 (Web Text))

Max Activating Token Index: 440

Text #16

Max Range: 5.4355. Min Range: -5.4355

Max Act: 4.9980. Min Act: -0.1700

Data Index: 238961 (C4 (Web Text))

Max Activating Token Index: 917

Text #17

Max Range: 5.4355. Min Range: -5.4355

Max Act: 5.0292. Min Act: -0.1700

Data Index: 1082038 (C4 (Web Text))

Max Activating Token Index: 733

Text #18

Max Range: 5.4355. Min Range: -5.4355

Max Act: 4.8730. Min Act: -0.1700

Data Index: 80739 (C4 (Web Text))

Max Activating Token Index: 498

Text #19

Max Range: 5.4355. Min Range: -5.4355

Max Act: 4.9980. Min Act: -0.1700

Data Index: 149247 (C4 (Web Text))

Max Activating Token Index: 83