Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 641 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.3495. Min Act: -0.1700

Data Index: 1091064 (C4 (Web Text))

Max Activating Token Index: 414

Text #1

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.2557. Min Act: -0.1700

Data Index: 1266817 (C4 (Web Text))

Max Activating Token Index: 485

Text #2

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.1307. Min Act: -0.1700

Data Index: 691554 (C4 (Web Text))

Max Activating Token Index: 728

Text #3

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.1620. Min Act: -0.1698

Data Index: 1302507 (C4 (Web Text))

Max Activating Token Index: 229

Text #4

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.0995. Min Act: -0.1699

Data Index: 1099545 (C4 (Web Text))

Max Activating Token Index: 220

Text #5

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.0682. Min Act: -0.1700

Data Index: 922249 (C4 (Web Text))

Max Activating Token Index: 661

Text #6

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.1307. Min Act: -0.1700

Data Index: 1290879 (C4 (Web Text))

Max Activating Token Index: 251

Text #7

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.0995. Min Act: -0.1700

Data Index: 960463 (C4 (Web Text))

Max Activating Token Index: 162

Text #8

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.1307. Min Act: -0.1700

Data Index: 24160 (C4 (Web Text))

Max Activating Token Index: 851

Text #9

Max Range: 5.3495. Min Range: -5.3495

Max Act: 5.0057. Min Act: -0.1700

Data Index: 495183 (C4 (Web Text))

Max Activating Token Index: 614

Text #10

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.9120. Min Act: -0.1700

Data Index: 431236 (C4 (Web Text))

Max Activating Token Index: 341

Text #11

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.9432. Min Act: -0.1700

Data Index: 447875 (C4 (Web Text))

Max Activating Token Index: 166

Text #12

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.9120. Min Act: -0.1699

Data Index: 232701 (C4 (Web Text))

Max Activating Token Index: 933

Text #13

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.9120. Min Act: -0.1700

Data Index: 430868 (C4 (Web Text))

Max Activating Token Index: 520

Text #14

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8807. Min Act: -0.1700

Data Index: 183964 (C4 (Web Text))

Max Activating Token Index: 591

Text #15

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8495. Min Act: -0.1700

Data Index: 279913 (C4 (Web Text))

Max Activating Token Index: 490

Text #16

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8807. Min Act: -0.1698

Data Index: 1021190 (C4 (Web Text))

Max Activating Token Index: 421

Text #17

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8807. Min Act: -0.1700

Data Index: 1023525 (C4 (Web Text))

Max Activating Token Index: 872

Text #18

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8495. Min Act: -0.1700

Data Index: 665060 (C4 (Web Text))

Max Activating Token Index: 716

Text #19

Max Range: 5.3495. Min Range: -5.3495

Max Act: 4.8495. Min Act: -0.1700

Data Index: 1057810 (C4 (Web Text))

Max Activating Token Index: 468