Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1116 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.2981. Min Act: -0.1573

Data Index: 277558 (C4 (Web Text))

Max Activating Token Index: 296

Text #1

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.3294. Min Act: -0.1700

Data Index: 1388462 (Python Code)

Max Activating Token Index: 152

Text #2

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.2981. Min Act: -0.1698

Data Index: 160358 (C4 (Web Text))

Max Activating Token Index: 230

Text #3

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.2669. Min Act: -0.1700

Data Index: 875493 (C4 (Web Text))

Max Activating Token Index: 39

Text #4

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.2356. Min Act: -0.1695

Data Index: 158148 (C4 (Web Text))

Max Activating Token Index: 320

Text #5

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.2044. Min Act: -0.1553

Data Index: 1010130 (C4 (Web Text))

Max Activating Token Index: 233

Text #6

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1731. Min Act: -0.1700

Data Index: 150757 (C4 (Web Text))

Max Activating Token Index: 816

Text #7

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1731. Min Act: -0.1700

Data Index: 1200778 (C4 (Web Text))

Max Activating Token Index: 1008

Text #8

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1731. Min Act: -0.1698

Data Index: 182562 (C4 (Web Text))

Max Activating Token Index: 183

Text #9

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1106. Min Act: -0.1700

Data Index: 1452684 (Python Code)

Max Activating Token Index: 26

Text #10

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1731. Min Act: -0.1651

Data Index: 1100813 (C4 (Web Text))

Max Activating Token Index: 444

Text #11

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.1106. Min Act: -0.1698

Data Index: 359499 (C4 (Web Text))

Max Activating Token Index: 640

Text #12

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.0481. Min Act: -0.1700

Data Index: 1213235 (C4 (Web Text))

Max Activating Token Index: 337

Text #13

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.0794. Min Act: -0.1681

Data Index: 506343 (C4 (Web Text))

Max Activating Token Index: 712

Text #14

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.0794. Min Act: -0.1699

Data Index: 1432260 (Python Code)

Max Activating Token Index: 335

Text #15

Max Range: 5.3294. Min Range: -5.3294

Max Act: 4.9856. Min Act: -0.1700

Data Index: 1414053 (Python Code)

Max Activating Token Index: 715

Text #16

Max Range: 5.3294. Min Range: -5.3294

Max Act: 4.9856. Min Act: -0.1178

Data Index: 1413696 (Python Code)

Max Activating Token Index: 916

Text #17

Max Range: 5.3294. Min Range: -5.3294

Max Act: 4.9856. Min Act: -0.1700

Data Index: 1602995 (Python Code)

Max Activating Token Index: 646

Text #18

Max Range: 5.3294. Min Range: -5.3294

Max Act: 5.0169. Min Act: -0.1700

Data Index: 204484 (C4 (Web Text))

Max Activating Token Index: 532

Text #19

Max Range: 5.3294. Min Range: -5.3294

Max Act: 4.9856. Min Act: -0.1688

Data Index: 1272326 (C4 (Web Text))

Max Activating Token Index: 613