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

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

Neuron 235 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: 5.7085. Min Range: -5.7085

Max Act: 5.7085. Min Act: -0.1700

Data Index: 856419 (C4 (Web Text))

Max Activating Token Index: 763

Text #1

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.6147. Min Act: -0.1700

Data Index: 1140518 (C4 (Web Text))

Max Activating Token Index: 552

Text #2

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.4897. Min Act: -0.1700

Data Index: 651668 (C4 (Web Text))

Max Activating Token Index: 600

Text #3

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.4272. Min Act: -0.1700

Data Index: 606602 (C4 (Web Text))

Max Activating Token Index: 423

Text #4

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.3960. Min Act: -0.1700

Data Index: 1088343 (C4 (Web Text))

Max Activating Token Index: 743

Text #5

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.3647. Min Act: -0.1700

Data Index: 757334 (C4 (Web Text))

Max Activating Token Index: 466

Text #6

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.3335. Min Act: -0.1700

Data Index: 450497 (C4 (Web Text))

Max Activating Token Index: 840

Text #7

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2085. Min Act: -0.1700

Data Index: 565164 (C4 (Web Text))

Max Activating Token Index: 783

Text #8

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2710. Min Act: -0.1700

Data Index: 1188648 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2397. Min Act: -0.1700

Data Index: 1253414 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2710. Min Act: -0.1700

Data Index: 997067 (C4 (Web Text))

Max Activating Token Index: 833

Text #11

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2397. Min Act: -0.1700

Data Index: 938688 (C4 (Web Text))

Max Activating Token Index: 492

Text #12

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.1772. Min Act: -0.1700

Data Index: 227908 (C4 (Web Text))

Max Activating Token Index: 401

Text #13

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2085. Min Act: -0.1700

Data Index: 196134 (C4 (Web Text))

Max Activating Token Index: 134

Text #14

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.1772. Min Act: -0.1700

Data Index: 193308 (C4 (Web Text))

Max Activating Token Index: 964

Text #15

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2085. Min Act: -0.1700

Data Index: 726328 (C4 (Web Text))

Max Activating Token Index: 494

Text #16

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2085. Min Act: -0.1700

Data Index: 1227414 (C4 (Web Text))

Max Activating Token Index: 101

Text #17

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.2085. Min Act: -0.1700

Data Index: 106282 (C4 (Web Text))

Max Activating Token Index: 814

Text #18

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.1460. Min Act: -0.1700

Data Index: 505894 (C4 (Web Text))

Max Activating Token Index: 612

Text #19

Max Range: 5.7085. Min Range: -5.7085

Max Act: 5.1460. Min Act: -0.1700

Data Index: 951478 (C4 (Web Text))

Max Activating Token Index: 832