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

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

Neuron 85 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: 5.2931. Min Range: -5.2931

Max Act: 5.2931. Min Act: -0.1700

Data Index: 1113117 (C4 (Web Text))

Max Activating Token Index: 289

Text #1

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.2306. Min Act: -0.1700

Data Index: 232385 (C4 (Web Text))

Max Activating Token Index: 457

Text #2

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1993. Min Act: -0.1700

Data Index: 118151 (C4 (Web Text))

Max Activating Token Index: 532

Text #3

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1056. Min Act: -0.1700

Data Index: 1050887 (C4 (Web Text))

Max Activating Token Index: 308

Text #4

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1681. Min Act: -0.1700

Data Index: 117124 (C4 (Web Text))

Max Activating Token Index: 552

Text #5

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1681. Min Act: -0.1700

Data Index: 1031308 (C4 (Web Text))

Max Activating Token Index: 828

Text #6

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1368. Min Act: -0.1700

Data Index: 637685 (C4 (Web Text))

Max Activating Token Index: 97

Text #7

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1368. Min Act: -0.1700

Data Index: 157157 (C4 (Web Text))

Max Activating Token Index: 486

Text #8

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1368. Min Act: -0.1700

Data Index: 1331030 (C4 (Web Text))

Max Activating Token Index: 325

Text #9

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1368. Min Act: -0.1700

Data Index: 41437 (C4 (Web Text))

Max Activating Token Index: 893

Text #10

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0743. Min Act: -0.1700

Data Index: 610474 (C4 (Web Text))

Max Activating Token Index: 47

Text #11

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0431. Min Act: -0.1700

Data Index: 1264407 (C4 (Web Text))

Max Activating Token Index: 568

Text #12

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0274. Min Act: -0.1700

Data Index: 576748 (C4 (Web Text))

Max Activating Token Index: 59

Text #13

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0743. Min Act: -0.1700

Data Index: 335616 (C4 (Web Text))

Max Activating Token Index: 299

Text #14

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0743. Min Act: -0.1700

Data Index: 4072 (C4 (Web Text))

Max Activating Token Index: 737

Text #15

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0743. Min Act: -0.1700

Data Index: 257887 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.1056. Min Act: -0.1700

Data Index: 18429 (C4 (Web Text))

Max Activating Token Index: 376

Text #17

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0274. Min Act: -0.1700

Data Index: 586451 (C4 (Web Text))

Max Activating Token Index: 549

Text #18

Max Range: 5.2931. Min Range: -5.2931

Max Act: 4.9806. Min Act: -0.1700

Data Index: 448148 (C4 (Web Text))

Max Activating Token Index: 205

Text #19

Max Range: 5.2931. Min Range: -5.2931

Max Act: 5.0431. Min Act: -0.1700

Data Index: 158441 (C4 (Web Text))

Max Activating Token Index: 54