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

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

Neuron 2009 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: 3.7222. Min Range: -3.7222

Max Act: 3.6909. Min Act: -0.1700

Data Index: 498753 (C4 (Web Text))

Max Activating Token Index: 396

Text #1

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6909. Min Act: -0.1700

Data Index: 244449 (C4 (Web Text))

Max Activating Token Index: 537

Text #2

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.7222. Min Act: -0.1700

Data Index: 920659 (C4 (Web Text))

Max Activating Token Index: 947

Text #3

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6596. Min Act: -0.1700

Data Index: 348599 (C4 (Web Text))

Max Activating Token Index: 450

Text #4

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6283. Min Act: -0.1700

Data Index: 791027 (C4 (Web Text))

Max Activating Token Index: 608

Text #5

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6283. Min Act: -0.1700

Data Index: 606233 (C4 (Web Text))

Max Activating Token Index: 670

Text #6

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6126. Min Act: -0.1700

Data Index: 581882 (C4 (Web Text))

Max Activating Token Index: 435

Text #7

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6909. Min Act: -0.1700

Data Index: 1015607 (C4 (Web Text))

Max Activating Token Index: 981

Text #8

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5657. Min Act: -0.1700

Data Index: 179959 (C4 (Web Text))

Max Activating Token Index: 667

Text #9

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6439. Min Act: -0.1700

Data Index: 132724 (C4 (Web Text))

Max Activating Token Index: 1003

Text #10

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5970. Min Act: -0.1700

Data Index: 774041 (C4 (Web Text))

Max Activating Token Index: 594

Text #11

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6909. Min Act: -0.1700

Data Index: 1089213 (C4 (Web Text))

Max Activating Token Index: 424

Text #12

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5813. Min Act: -0.1700

Data Index: 439986 (C4 (Web Text))

Max Activating Token Index: 1018

Text #13

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5813. Min Act: -0.1700

Data Index: 770560 (C4 (Web Text))

Max Activating Token Index: 994

Text #14

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5813. Min Act: -0.1700

Data Index: 63630 (C4 (Web Text))

Max Activating Token Index: 615

Text #15

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5970. Min Act: -0.1700

Data Index: 1005792 (C4 (Web Text))

Max Activating Token Index: 942

Text #16

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5970. Min Act: -0.1700

Data Index: 470954 (C4 (Web Text))

Max Activating Token Index: 360

Text #17

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5813. Min Act: -0.1700

Data Index: 572914 (C4 (Web Text))

Max Activating Token Index: 878

Text #18

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.6596. Min Act: -0.1700

Data Index: 735442 (C4 (Web Text))

Max Activating Token Index: 625

Text #19

Max Range: 3.7222. Min Range: -3.7222

Max Act: 3.5500. Min Act: -0.1700

Data Index: 44557 (C4 (Web Text))

Max Activating Token Index: 592