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

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

Neuron 1603 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: 5.0110. Min Range: -5.0110

Max Act: 5.0110. Min Act: -0.1700

Data Index: 1001007 (C4 (Web Text))

Max Activating Token Index: 629

Text #1

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.8548. Min Act: -0.1700

Data Index: 588029 (C4 (Web Text))

Max Activating Token Index: 826

Text #2

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.8235. Min Act: -0.1700

Data Index: 35166 (C4 (Web Text))

Max Activating Token Index: 744

Text #3

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7923. Min Act: -0.1700

Data Index: 775435 (C4 (Web Text))

Max Activating Token Index: 469

Text #4

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7610. Min Act: -0.1698

Data Index: 867273 (C4 (Web Text))

Max Activating Token Index: 822

Text #5

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7610. Min Act: -0.1699

Data Index: 1140224 (C4 (Web Text))

Max Activating Token Index: 656

Text #6

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7923. Min Act: -0.1655

Data Index: 728278 (C4 (Web Text))

Max Activating Token Index: 763

Text #7

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7610. Min Act: -0.1699

Data Index: 392374 (C4 (Web Text))

Max Activating Token Index: 288

Text #8

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7610. Min Act: -0.1700

Data Index: 1198342 (C4 (Web Text))

Max Activating Token Index: 208

Text #9

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.7298. Min Act: -0.1700

Data Index: 509276 (C4 (Web Text))

Max Activating Token Index: 329

Text #10

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6985. Min Act: -0.1700

Data Index: 463434 (C4 (Web Text))

Max Activating Token Index: 428

Text #11

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6673. Min Act: -0.1700

Data Index: 163856 (C4 (Web Text))

Max Activating Token Index: 933

Text #12

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6048. Min Act: -0.1699

Data Index: 622520 (C4 (Web Text))

Max Activating Token Index: 661

Text #13

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6048. Min Act: -0.1700

Data Index: 181434 (C4 (Web Text))

Max Activating Token Index: 459

Text #14

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6360. Min Act: -0.1700

Data Index: 288643 (C4 (Web Text))

Max Activating Token Index: 158

Text #15

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6360. Min Act: -0.1700

Data Index: 840903 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.6360. Min Act: -0.1679

Data Index: 1143211 (C4 (Web Text))

Max Activating Token Index: 465

Text #17

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.5735. Min Act: -0.1700

Data Index: 612281 (C4 (Web Text))

Max Activating Token Index: 965

Text #18

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.5735. Min Act: -0.1700

Data Index: 823794 (C4 (Web Text))

Max Activating Token Index: 433

Text #19

Max Range: 5.0110. Min Range: -5.0110

Max Act: 4.5423. Min Act: -0.1700

Data Index: 860819 (C4 (Web Text))

Max Activating Token Index: 385