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

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

Neuron 1072 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: 3.7453. Min Range: -3.7453

Max Act: 3.7453. Min Act: -0.1700

Data Index: 613111 (C4 (Web Text))

Max Activating Token Index: 363

Text #1

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.7140. Min Act: -0.1700

Data Index: 391353 (C4 (Web Text))

Max Activating Token Index: 747

Text #2

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.7297. Min Act: -0.1700

Data Index: 225986 (C4 (Web Text))

Max Activating Token Index: 1011

Text #3

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.7140. Min Act: -0.1700

Data Index: 571754 (C4 (Web Text))

Max Activating Token Index: 55

Text #4

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6827. Min Act: -0.1700

Data Index: 200469 (C4 (Web Text))

Max Activating Token Index: 418

Text #5

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6984. Min Act: -0.1700

Data Index: 850971 (C4 (Web Text))

Max Activating Token Index: 159

Text #6

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6827. Min Act: -0.1700

Data Index: 371954 (C4 (Web Text))

Max Activating Token Index: 227

Text #7

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6514. Min Act: -0.1700

Data Index: 1246401 (C4 (Web Text))

Max Activating Token Index: 112

Text #8

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6671. Min Act: -0.1700

Data Index: 384239 (C4 (Web Text))

Max Activating Token Index: 639

Text #9

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6827. Min Act: -0.1700

Data Index: 699338 (C4 (Web Text))

Max Activating Token Index: 787

Text #10

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6671. Min Act: -0.1700

Data Index: 1064571 (C4 (Web Text))

Max Activating Token Index: 653

Text #11

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6358. Min Act: -0.1700

Data Index: 301474 (C4 (Web Text))

Max Activating Token Index: 382

Text #12

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6201. Min Act: -0.1700

Data Index: 1316307 (C4 (Web Text))

Max Activating Token Index: 524

Text #13

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6514. Min Act: -0.1700

Data Index: 124574 (C4 (Web Text))

Max Activating Token Index: 922

Text #14

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6045. Min Act: -0.1700

Data Index: 316470 (C4 (Web Text))

Max Activating Token Index: 176

Text #15

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6514. Min Act: -0.1700

Data Index: 1065221 (C4 (Web Text))

Max Activating Token Index: 720

Text #16

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6201. Min Act: -0.1700

Data Index: 297366 (C4 (Web Text))

Max Activating Token Index: 76

Text #17

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6514. Min Act: -0.1700

Data Index: 452430 (C4 (Web Text))

Max Activating Token Index: 853

Text #18

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6045. Min Act: -0.1700

Data Index: 910395 (C4 (Web Text))

Max Activating Token Index: 722

Text #19

Max Range: 3.7453. Min Range: -3.7453

Max Act: 3.6671. Min Act: -0.1700

Data Index: 414100 (C4 (Web Text))

Max Activating Token Index: 502