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

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

Neuron 2006 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: 3.3839. Min Range: -3.3839

Max Act: 3.3839. Min Act: -0.1700

Data Index: 1132162 (C4 (Web Text))

Max Activating Token Index: 91

Text #1

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2740. Min Act: -0.1700

Data Index: 248065 (C4 (Web Text))

Max Activating Token Index: 72

Text #2

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2898. Min Act: -0.1700

Data Index: 92681 (C4 (Web Text))

Max Activating Token Index: 241

Text #3

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2740. Min Act: -0.1700

Data Index: 454711 (C4 (Web Text))

Max Activating Token Index: 707

Text #4

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2583. Min Act: -0.1700

Data Index: 157582 (C4 (Web Text))

Max Activating Token Index: 616

Text #5

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2112. Min Act: -0.1700

Data Index: 142917 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2112. Min Act: -0.1700

Data Index: 964788 (C4 (Web Text))

Max Activating Token Index: 162

Text #7

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1954. Min Act: -0.1700

Data Index: 499820 (C4 (Web Text))

Max Activating Token Index: 715

Text #8

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.2112. Min Act: -0.1700

Data Index: 142210 (C4 (Web Text))

Max Activating Token Index: 777

Text #9

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1639. Min Act: -0.1700

Data Index: 1275301 (C4 (Web Text))

Max Activating Token Index: 689

Text #10

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1954. Min Act: -0.1700

Data Index: 258586 (C4 (Web Text))

Max Activating Token Index: 639

Text #11

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1797. Min Act: -0.1700

Data Index: 336239 (C4 (Web Text))

Max Activating Token Index: 560

Text #12

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1797. Min Act: -0.1700

Data Index: 324273 (C4 (Web Text))

Max Activating Token Index: 826

Text #13

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1797. Min Act: -0.1700

Data Index: 662223 (C4 (Web Text))

Max Activating Token Index: 625

Text #14

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1954. Min Act: -0.1700

Data Index: 11830 (C4 (Web Text))

Max Activating Token Index: 529

Text #15

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1797. Min Act: -0.1700

Data Index: 466091 (C4 (Web Text))

Max Activating Token Index: 102

Text #16

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1639. Min Act: -0.1700

Data Index: 24522 (C4 (Web Text))

Max Activating Token Index: 657

Text #17

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1954. Min Act: -0.1700

Data Index: 869151 (C4 (Web Text))

Max Activating Token Index: 121

Text #18

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1797. Min Act: -0.1700

Data Index: 555204 (C4 (Web Text))

Max Activating Token Index: 1003

Text #19

Max Range: 3.3839. Min Range: -3.3839

Max Act: 3.1639. Min Act: -0.1700

Data Index: 1020810 (C4 (Web Text))

Max Activating Token Index: 590