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

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

Neuron 725 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: 5.6840. Min Range: -5.6840

Max Act: 5.6840. Min Act: -0.1700

Data Index: 1297629 (C4 (Web Text))

Max Activating Token Index: 80

Text #1

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.6215. Min Act: -0.1700

Data Index: 165092 (C4 (Web Text))

Max Activating Token Index: 156

Text #2

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.5590. Min Act: -0.1700

Data Index: 578015 (C4 (Web Text))

Max Activating Token Index: 119

Text #3

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4965. Min Act: -0.1700

Data Index: 1109847 (C4 (Web Text))

Max Activating Token Index: 228

Text #4

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4965. Min Act: -0.1700

Data Index: 537324 (C4 (Web Text))

Max Activating Token Index: 184

Text #5

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4340. Min Act: -0.1700

Data Index: 1177943 (C4 (Web Text))

Max Activating Token Index: 491

Text #6

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4653. Min Act: -0.1700

Data Index: 538625 (C4 (Web Text))

Max Activating Token Index: 179

Text #7

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4028. Min Act: -0.1700

Data Index: 787202 (C4 (Web Text))

Max Activating Token Index: 288

Text #8

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4028. Min Act: -0.1700

Data Index: 665292 (C4 (Web Text))

Max Activating Token Index: 181

Text #9

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.4028. Min Act: -0.1700

Data Index: 1314271 (C4 (Web Text))

Max Activating Token Index: 478

Text #10

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3403. Min Act: -0.1700

Data Index: 147749 (C4 (Web Text))

Max Activating Token Index: 250

Text #11

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3403. Min Act: -0.1700

Data Index: 458607 (C4 (Web Text))

Max Activating Token Index: 382

Text #12

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3090. Min Act: -0.1700

Data Index: 923777 (C4 (Web Text))

Max Activating Token Index: 566

Text #13

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3715. Min Act: -0.1700

Data Index: 142011 (C4 (Web Text))

Max Activating Token Index: 450

Text #14

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3715. Min Act: -0.1700

Data Index: 1274154 (C4 (Web Text))

Max Activating Token Index: 870

Text #15

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3090. Min Act: -0.1700

Data Index: 1347723 (C4 (Web Text))

Max Activating Token Index: 696

Text #16

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3403. Min Act: -0.1700

Data Index: 840051 (C4 (Web Text))

Max Activating Token Index: 194

Text #17

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3090. Min Act: -0.1700

Data Index: 845129 (C4 (Web Text))

Max Activating Token Index: 498

Text #18

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.2778. Min Act: -0.1700

Data Index: 98479 (C4 (Web Text))

Max Activating Token Index: 116

Text #19

Max Range: 5.6840. Min Range: -5.6840

Max Act: 5.3090. Min Act: -0.1700

Data Index: 857986 (C4 (Web Text))

Max Activating Token Index: 647