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

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

Neuron 1373 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.5054. Min Act: -0.1700

Data Index: 1444536 (Python Code)

Max Activating Token Index: 748

Text #1

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4584. Min Act: -0.1700

Data Index: 974877 (C4 (Web Text))

Max Activating Token Index: 863

Text #2

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4114. Min Act: -0.1700

Data Index: 1037815 (C4 (Web Text))

Max Activating Token Index: 874

Text #3

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4271. Min Act: -0.1700

Data Index: 1420555 (Python Code)

Max Activating Token Index: 217

Text #4

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4271. Min Act: -0.1700

Data Index: 1206525 (C4 (Web Text))

Max Activating Token Index: 978

Text #5

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4114. Min Act: -0.1700

Data Index: 583225 (C4 (Web Text))

Max Activating Token Index: 8

Text #6

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.4114. Min Act: -0.1700

Data Index: 1061958 (C4 (Web Text))

Max Activating Token Index: 439

Text #7

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3957. Min Act: -0.1700

Data Index: 190442 (C4 (Web Text))

Max Activating Token Index: 981

Text #8

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3957. Min Act: -0.1700

Data Index: 1194983 (C4 (Web Text))

Max Activating Token Index: 791

Text #9

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3957. Min Act: -0.1700

Data Index: 811733 (C4 (Web Text))

Max Activating Token Index: 752

Text #10

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3957. Min Act: -0.1700

Data Index: 742931 (C4 (Web Text))

Max Activating Token Index: 489

Text #11

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3957. Min Act: -0.1700

Data Index: 982601 (C4 (Web Text))

Max Activating Token Index: 1013

Text #12

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3800. Min Act: -0.1700

Data Index: 566183 (C4 (Web Text))

Max Activating Token Index: 54

Text #13

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3800. Min Act: -0.1700

Data Index: 31936 (C4 (Web Text))

Max Activating Token Index: 72

Text #14

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3643. Min Act: -0.1700

Data Index: 1419003 (Python Code)

Max Activating Token Index: 792

Text #15

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3643. Min Act: -0.1700

Data Index: 931540 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3800. Min Act: -0.1700

Data Index: 891763 (C4 (Web Text))

Max Activating Token Index: 52

Text #17

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3643. Min Act: -0.1700

Data Index: 451085 (C4 (Web Text))

Max Activating Token Index: 296

Text #18

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3486. Min Act: -0.1700

Data Index: 1032491 (C4 (Web Text))

Max Activating Token Index: 627

Text #19

Max Range: 3.5054. Min Range: -3.5054

Max Act: 3.3486. Min Act: -0.1700

Data Index: 467192 (C4 (Web Text))

Max Activating Token Index: 550