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

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

Neuron 1066 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: 5.2023. Min Range: -5.2023

Max Act: 5.2023. Min Act: -0.1700

Data Index: 988075 (C4 (Web Text))

Max Activating Token Index: 72

Text #1

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.1710. Min Act: -0.1700

Data Index: 923813 (C4 (Web Text))

Max Activating Token Index: 149

Text #2

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.1242. Min Act: -0.1700

Data Index: 37654 (C4 (Web Text))

Max Activating Token Index: 110

Text #3

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.1867. Min Act: -0.1700

Data Index: 318419 (C4 (Web Text))

Max Activating Token Index: 573

Text #4

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0773. Min Act: -0.1700

Data Index: 464099 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0617. Min Act: -0.1700

Data Index: 1060334 (C4 (Web Text))

Max Activating Token Index: 113

Text #6

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.1710. Min Act: -0.1700

Data Index: 45169 (C4 (Web Text))

Max Activating Token Index: 86

Text #7

Max Range: 5.2023. Min Range: -5.2023

Max Act: 4.9367. Min Act: -0.1700

Data Index: 961980 (C4 (Web Text))

Max Activating Token Index: 407

Text #8

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0929. Min Act: -0.1700

Data Index: 565126 (C4 (Web Text))

Max Activating Token Index: 832

Text #9

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.1242. Min Act: -0.1700

Data Index: 177404 (C4 (Web Text))

Max Activating Token Index: 70

Text #10

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0773. Min Act: -0.1700

Data Index: 675077 (C4 (Web Text))

Max Activating Token Index: 650

Text #11

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0929. Min Act: -0.1700

Data Index: 809445 (C4 (Web Text))

Max Activating Token Index: 620

Text #12

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0773. Min Act: -0.1700

Data Index: 1494749 (Python Code)

Max Activating Token Index: 808

Text #13

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0460. Min Act: -0.1700

Data Index: 621015 (C4 (Web Text))

Max Activating Token Index: 733

Text #14

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0304. Min Act: -0.1700

Data Index: 981597 (C4 (Web Text))

Max Activating Token Index: 261

Text #15

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0304. Min Act: -0.1700

Data Index: 1101420 (C4 (Web Text))

Max Activating Token Index: 75

Text #16

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0773. Min Act: -0.1700

Data Index: 773841 (C4 (Web Text))

Max Activating Token Index: 407

Text #17

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0617. Min Act: -0.1700

Data Index: 589204 (C4 (Web Text))

Max Activating Token Index: 808

Text #18

Max Range: 5.2023. Min Range: -5.2023

Max Act: 4.9523. Min Act: -0.1700

Data Index: 600055 (C4 (Web Text))

Max Activating Token Index: 128

Text #19

Max Range: 5.2023. Min Range: -5.2023

Max Act: 5.0148. Min Act: -0.1700

Data Index: 42789 (C4 (Web Text))

Max Activating Token Index: 964