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

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

Neuron 159 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7843. Min Act: -0.1700

Data Index: 1000166 (C4 (Web Text))

Max Activating Token Index: 713

Text #1

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7530. Min Act: -0.1700

Data Index: 1163719 (C4 (Web Text))

Max Activating Token Index: 473

Text #2

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7530. Min Act: -0.1700

Data Index: 873055 (C4 (Web Text))

Max Activating Token Index: 598

Text #3

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6904. Min Act: -0.1700

Data Index: 614685 (C4 (Web Text))

Max Activating Token Index: 576

Text #4

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7217. Min Act: -0.1700

Data Index: 162771 (C4 (Web Text))

Max Activating Token Index: 755

Text #5

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6904. Min Act: -0.1700

Data Index: 892638 (C4 (Web Text))

Max Activating Token Index: 648

Text #6

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7217. Min Act: -0.1700

Data Index: 157348 (C4 (Web Text))

Max Activating Token Index: 583

Text #7

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6904. Min Act: -0.1700

Data Index: 118999 (C4 (Web Text))

Max Activating Token Index: 289

Text #8

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.7217. Min Act: -0.1700

Data Index: 609845 (C4 (Web Text))

Max Activating Token Index: 754

Text #9

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 869923 (C4 (Web Text))

Max Activating Token Index: 886

Text #10

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 414308 (C4 (Web Text))

Max Activating Token Index: 863

Text #11

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6278. Min Act: -0.1700

Data Index: 179738 (C4 (Web Text))

Max Activating Token Index: 688

Text #12

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 275464 (C4 (Web Text))

Max Activating Token Index: 900

Text #13

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 106571 (C4 (Web Text))

Max Activating Token Index: 29

Text #14

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 204212 (C4 (Web Text))

Max Activating Token Index: 178

Text #15

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6591. Min Act: -0.1700

Data Index: 196331 (C4 (Web Text))

Max Activating Token Index: 621

Text #16

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6278. Min Act: -0.1700

Data Index: 528035 (C4 (Web Text))

Max Activating Token Index: 276

Text #17

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6278. Min Act: -0.1700

Data Index: 30556 (C4 (Web Text))

Max Activating Token Index: 780

Text #18

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.6278. Min Act: -0.1700

Data Index: 1111091 (C4 (Web Text))

Max Activating Token Index: 620

Text #19

Max Range: 3.7843. Min Range: -3.7843

Max Act: 3.5965. Min Act: -0.1700

Data Index: 986112 (C4 (Web Text))

Max Activating Token Index: 793