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

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

Neuron 88 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.2963. Min Act: -0.1700

Data Index: 106021 (C4 (Web Text))

Max Activating Token Index: 939

Text #1

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.2963. Min Act: -0.1700

Data Index: 443119 (C4 (Web Text))

Max Activating Token Index: 92

Text #2

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.2963. Min Act: -0.1700

Data Index: 230425 (C4 (Web Text))

Max Activating Token Index: 948

Text #3

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.2650. Min Act: -0.1700

Data Index: 1220605 (C4 (Web Text))

Max Activating Token Index: 1005

Text #4

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.2338. Min Act: -0.1700

Data Index: 1259825 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1713. Min Act: -0.1700

Data Index: 1197237 (C4 (Web Text))

Max Activating Token Index: 915

Text #6

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1713. Min Act: -0.1700

Data Index: 498390 (C4 (Web Text))

Max Activating Token Index: 492

Text #7

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1713. Min Act: -0.1700

Data Index: 413460 (C4 (Web Text))

Max Activating Token Index: 303

Text #8

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1088. Min Act: -0.1700

Data Index: 908318 (C4 (Web Text))

Max Activating Token Index: 115

Text #9

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1088. Min Act: -0.1700

Data Index: 179390 (C4 (Web Text))

Max Activating Token Index: 669

Text #10

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 386379 (C4 (Web Text))

Max Activating Token Index: 559

Text #11

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1088. Min Act: -0.1700

Data Index: 866401 (C4 (Web Text))

Max Activating Token Index: 220

Text #12

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1088. Min Act: -0.1700

Data Index: 918662 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 412000 (C4 (Web Text))

Max Activating Token Index: 266

Text #14

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 746601 (C4 (Web Text))

Max Activating Token Index: 127

Text #15

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 1291278 (C4 (Web Text))

Max Activating Token Index: 839

Text #16

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 180390 (C4 (Web Text))

Max Activating Token Index: 850

Text #17

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 1600 (C4 (Web Text))

Max Activating Token Index: 143

Text #18

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.1088. Min Act: -0.1700

Data Index: 968162 (C4 (Web Text))

Max Activating Token Index: 154

Text #19

Max Range: 5.2963. Min Range: -5.2963

Max Act: 5.0775. Min Act: -0.1700

Data Index: 230426 (C4 (Web Text))

Max Activating Token Index: 132