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

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

Neuron 18 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.7596. Min Act: -0.1700

Data Index: 869470 (C4 (Web Text))

Max Activating Token Index: 41

Text #1

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.7127. Min Act: -0.1700

Data Index: 439511 (C4 (Web Text))

Max Activating Token Index: 24

Text #2

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.7283. Min Act: -0.1700

Data Index: 632191 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6814. Min Act: -0.1700

Data Index: 445607 (C4 (Web Text))

Max Activating Token Index: 877

Text #4

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6970. Min Act: -0.1700

Data Index: 706398 (C4 (Web Text))

Max Activating Token Index: 385

Text #5

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6657. Min Act: -0.1700

Data Index: 1203157 (C4 (Web Text))

Max Activating Token Index: 725

Text #6

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6657. Min Act: -0.1700

Data Index: 1000573 (C4 (Web Text))

Max Activating Token Index: 22

Text #7

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6344. Min Act: -0.1700

Data Index: 879832 (C4 (Web Text))

Max Activating Token Index: 791

Text #8

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6501. Min Act: -0.1700

Data Index: 682032 (C4 (Web Text))

Max Activating Token Index: 476

Text #9

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6344. Min Act: -0.1700

Data Index: 1028800 (C4 (Web Text))

Max Activating Token Index: 824

Text #10

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6188. Min Act: -0.1700

Data Index: 869069 (C4 (Web Text))

Max Activating Token Index: 8

Text #11

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5875. Min Act: -0.1700

Data Index: 143620 (C4 (Web Text))

Max Activating Token Index: 700

Text #12

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6031. Min Act: -0.1700

Data Index: 1163184 (C4 (Web Text))

Max Activating Token Index: 21

Text #13

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6031. Min Act: -0.1700

Data Index: 730429 (C4 (Web Text))

Max Activating Token Index: 625

Text #14

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.6031. Min Act: -0.1700

Data Index: 1169294 (C4 (Web Text))

Max Activating Token Index: 601

Text #15

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5875. Min Act: -0.1700

Data Index: 778784 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5718. Min Act: -0.1700

Data Index: 844868 (C4 (Web Text))

Max Activating Token Index: 676

Text #17

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5405. Min Act: -0.1700

Data Index: 412927 (C4 (Web Text))

Max Activating Token Index: 927

Text #18

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5561. Min Act: -0.1700

Data Index: 541655 (C4 (Web Text))

Max Activating Token Index: 51

Text #19

Max Range: 3.7596. Min Range: -3.7596

Max Act: 3.5561. Min Act: -0.1700

Data Index: 429642 (C4 (Web Text))

Max Activating Token Index: 692