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

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

Neuron 923 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: 7.7951. Min Range: -7.7951

Max Act: 7.7951. Min Act: -0.1700

Data Index: 1405873 (Python Code)

Max Activating Token Index: 1008

Text #1

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.6076. Min Act: -0.1700

Data Index: 839086 (C4 (Web Text))

Max Activating Token Index: 939

Text #2

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.7326. Min Act: -0.1699

Data Index: 1403156 (Python Code)

Max Activating Token Index: 974

Text #3

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.6076. Min Act: -0.1700

Data Index: 1472887 (Python Code)

Max Activating Token Index: 810

Text #4

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.4826. Min Act: -0.1700

Data Index: 1496561 (Python Code)

Max Activating Token Index: 851

Text #5

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.4201. Min Act: -0.1700

Data Index: 1511882 (Python Code)

Max Activating Token Index: 236

Text #6

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.4201. Min Act: -0.1700

Data Index: 65581 (C4 (Web Text))

Max Activating Token Index: 333

Text #7

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.4201. Min Act: -0.1700

Data Index: 1401331 (Python Code)

Max Activating Token Index: 245

Text #8

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2951. Min Act: -0.1700

Data Index: 257609 (C4 (Web Text))

Max Activating Token Index: 671

Text #9

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2326. Min Act: -0.1700

Data Index: 1618576 (Python Code)

Max Activating Token Index: 278

Text #10

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2326. Min Act: -0.1700

Data Index: 1573917 (Python Code)

Max Activating Token Index: 565

Text #11

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2951. Min Act: -0.1700

Data Index: 1643234 (Python Code)

Max Activating Token Index: 995

Text #12

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.1701. Min Act: -0.1700

Data Index: 1345747 (C4 (Web Text))

Max Activating Token Index: 781

Text #13

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2951. Min Act: -0.1700

Data Index: 1283975 (C4 (Web Text))

Max Activating Token Index: 701

Text #14

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2326. Min Act: -0.1700

Data Index: 457400 (C4 (Web Text))

Max Activating Token Index: 943

Text #15

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2326. Min Act: -0.1700

Data Index: 56023 (C4 (Web Text))

Max Activating Token Index: 954

Text #16

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.1076. Min Act: -0.1700

Data Index: 180828 (C4 (Web Text))

Max Activating Token Index: 678

Text #17

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.2326. Min Act: -0.1700

Data Index: 1577412 (Python Code)

Max Activating Token Index: 303

Text #18

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.1701. Min Act: -0.1700

Data Index: 1558863 (Python Code)

Max Activating Token Index: 357

Text #19

Max Range: 7.7951. Min Range: -7.7951

Max Act: 7.1701. Min Act: -0.1700

Data Index: 1423547 (Python Code)

Max Activating Token Index: 223