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

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

Neuron 1470 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: 5.3460. Min Range: -5.3460

Max Act: 5.3460. Min Act: -0.1700

Data Index: 889245 (C4 (Web Text))

Max Activating Token Index: 568

Text #1

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.3460. Min Act: -0.1700

Data Index: 65790 (C4 (Web Text))

Max Activating Token Index: 157

Text #2

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.3148. Min Act: -0.1700

Data Index: 144119 (C4 (Web Text))

Max Activating Token Index: 642

Text #3

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.2210. Min Act: -0.1700

Data Index: 889198 (C4 (Web Text))

Max Activating Token Index: 829

Text #4

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.2210. Min Act: -0.1700

Data Index: 509921 (C4 (Web Text))

Max Activating Token Index: 493

Text #5

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.2210. Min Act: -0.1700

Data Index: 933257 (C4 (Web Text))

Max Activating Token Index: 810

Text #6

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.1898. Min Act: -0.1700

Data Index: 794510 (C4 (Web Text))

Max Activating Token Index: 749

Text #7

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.1585. Min Act: -0.1700

Data Index: 50464 (C4 (Web Text))

Max Activating Token Index: 89

Text #8

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0960. Min Act: -0.1700

Data Index: 521787 (C4 (Web Text))

Max Activating Token Index: 825

Text #9

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0960. Min Act: -0.1700

Data Index: 887397 (C4 (Web Text))

Max Activating Token Index: 688

Text #10

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.1273. Min Act: -0.1700

Data Index: 755690 (C4 (Web Text))

Max Activating Token Index: 735

Text #11

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.1273. Min Act: -0.1700

Data Index: 19979 (C4 (Web Text))

Max Activating Token Index: 194

Text #12

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0960. Min Act: -0.1700

Data Index: 528025 (C4 (Web Text))

Max Activating Token Index: 916

Text #13

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.1273. Min Act: -0.1700

Data Index: 1044543 (C4 (Web Text))

Max Activating Token Index: 532

Text #14

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0960. Min Act: -0.1700

Data Index: 635623 (C4 (Web Text))

Max Activating Token Index: 781

Text #15

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0335. Min Act: -0.1700

Data Index: 798288 (C4 (Web Text))

Max Activating Token Index: 812

Text #16

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0335. Min Act: -0.1699

Data Index: 22526 (C4 (Web Text))

Max Activating Token Index: 436

Text #17

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0023. Min Act: -0.1700

Data Index: 1029981 (C4 (Web Text))

Max Activating Token Index: 810

Text #18

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0023. Min Act: -0.1700

Data Index: 97172 (C4 (Web Text))

Max Activating Token Index: 880

Text #19

Max Range: 5.3460. Min Range: -5.3460

Max Act: 5.0335. Min Act: -0.1700

Data Index: 741955 (C4 (Web Text))

Max Activating Token Index: 508