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

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

Neuron 1655 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: 5.9500. Min Range: -5.9500

Max Act: 5.9500. Min Act: -0.1700

Data Index: 913212 (C4 (Web Text))

Max Activating Token Index: 611

Text #1

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8563. Min Act: -0.1700

Data Index: 537854 (C4 (Web Text))

Max Activating Token Index: 348

Text #2

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8563. Min Act: -0.1699

Data Index: 175543 (C4 (Web Text))

Max Activating Token Index: 676

Text #3

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8563. Min Act: -0.1700

Data Index: 372686 (C4 (Web Text))

Max Activating Token Index: 392

Text #4

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8563. Min Act: -0.1700

Data Index: 1048279 (C4 (Web Text))

Max Activating Token Index: 315

Text #5

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8563. Min Act: -0.1699

Data Index: 1183391 (C4 (Web Text))

Max Activating Token Index: 681

Text #6

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7625. Min Act: -0.1700

Data Index: 110330 (C4 (Web Text))

Max Activating Token Index: 540

Text #7

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7000. Min Act: -0.1700

Data Index: 150531 (C4 (Web Text))

Max Activating Token Index: 612

Text #8

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.8250. Min Act: -0.1700

Data Index: 890112 (C4 (Web Text))

Max Activating Token Index: 521

Text #9

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.9188. Min Act: -0.1700

Data Index: 282655 (C4 (Web Text))

Max Activating Token Index: 869

Text #10

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7625. Min Act: -0.1700

Data Index: 1162222 (C4 (Web Text))

Max Activating Token Index: 616

Text #11

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7313. Min Act: -0.1700

Data Index: 815474 (C4 (Web Text))

Max Activating Token Index: 728

Text #12

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7313. Min Act: -0.1700

Data Index: 685918 (C4 (Web Text))

Max Activating Token Index: 420

Text #13

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7000. Min Act: -0.1699

Data Index: 274361 (C4 (Web Text))

Max Activating Token Index: 392

Text #14

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.6375. Min Act: -0.1700

Data Index: 508522 (C4 (Web Text))

Max Activating Token Index: 929

Text #15

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7000. Min Act: -0.1700

Data Index: 390468 (C4 (Web Text))

Max Activating Token Index: 274

Text #16

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7000. Min Act: -0.1700

Data Index: 533567 (C4 (Web Text))

Max Activating Token Index: 42

Text #17

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7938. Min Act: -0.1699

Data Index: 409084 (C4 (Web Text))

Max Activating Token Index: 974

Text #18

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.7000. Min Act: -0.1700

Data Index: 812699 (C4 (Web Text))

Max Activating Token Index: 861

Text #19

Max Range: 5.9500. Min Range: -5.9500

Max Act: 5.6063. Min Act: -0.1700

Data Index: 477810 (C4 (Web Text))

Max Activating Token Index: 752