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

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

Neuron 773 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1725. Min Range: -5.1725

Max Act: 5.1725. Min Act: -0.1700

Data Index: 249691 (C4 (Web Text))

Max Activating Token Index: 59

Text #1

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9850. Min Act: -0.1700

Data Index: 631850 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9537. Min Act: -0.1700

Data Index: 868144 (C4 (Web Text))

Max Activating Token Index: 645

Text #3

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9850. Min Act: -0.1700

Data Index: 1304075 (C4 (Web Text))

Max Activating Token Index: 40

Text #4

Max Range: 5.1725. Min Range: -5.1725

Max Act: 5.0162. Min Act: -0.1700

Data Index: 1302053 (C4 (Web Text))

Max Activating Token Index: 933

Text #5

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9537. Min Act: -0.1700

Data Index: 93214 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9225. Min Act: -0.1700

Data Index: 1310915 (C4 (Web Text))

Max Activating Token Index: 624

Text #7

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9225. Min Act: -0.1700

Data Index: 751018 (C4 (Web Text))

Max Activating Token Index: 514

Text #8

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.9537. Min Act: -0.1700

Data Index: 73963 (C4 (Web Text))

Max Activating Token Index: 352

Text #9

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8912. Min Act: -0.1700

Data Index: 307339 (C4 (Web Text))

Max Activating Token Index: 31

Text #10

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8912. Min Act: -0.1700

Data Index: 454041 (C4 (Web Text))

Max Activating Token Index: 553

Text #11

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8912. Min Act: -0.1700

Data Index: 353183 (C4 (Web Text))

Max Activating Token Index: 183

Text #12

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8600. Min Act: -0.1700

Data Index: 672255 (C4 (Web Text))

Max Activating Token Index: 54

Text #13

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8600. Min Act: -0.1700

Data Index: 405830 (C4 (Web Text))

Max Activating Token Index: 54

Text #14

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8287. Min Act: -0.1700

Data Index: 376692 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.8287. Min Act: -0.1700

Data Index: 159921 (C4 (Web Text))

Max Activating Token Index: 107

Text #16

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.7975. Min Act: -0.1700

Data Index: 520326 (C4 (Web Text))

Max Activating Token Index: 176

Text #17

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.7662. Min Act: -0.1700

Data Index: 466556 (C4 (Web Text))

Max Activating Token Index: 889

Text #18

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.6412. Min Act: -0.1700

Data Index: 805196 (C4 (Web Text))

Max Activating Token Index: 666

Text #19

Max Range: 5.1725. Min Range: -5.1725

Max Act: 4.7037. Min Act: -0.1700

Data Index: 528409 (C4 (Web Text))

Max Activating Token Index: 485