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

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

Neuron 607 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.8059. Min Act: -0.1700

Data Index: 1345608 (C4 (Web Text))

Max Activating Token Index: 773

Text #1

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.7277. Min Act: -0.1700

Data Index: 189039 (C4 (Web Text))

Max Activating Token Index: 736

Text #2

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6338. Min Act: -0.1700

Data Index: 1582695 (Python Code)

Max Activating Token Index: 130

Text #3

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6338. Min Act: -0.1700

Data Index: 1027576 (C4 (Web Text))

Max Activating Token Index: 730

Text #4

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6338. Min Act: -0.1700

Data Index: 1018954 (C4 (Web Text))

Max Activating Token Index: 917

Text #5

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6338. Min Act: -0.1700

Data Index: 236823 (C4 (Web Text))

Max Activating Token Index: 639

Text #6

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6181. Min Act: -0.1700

Data Index: 637385 (C4 (Web Text))

Max Activating Token Index: 800

Text #7

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6181. Min Act: -0.1700

Data Index: 226475 (C4 (Web Text))

Max Activating Token Index: 323

Text #8

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.5868. Min Act: -0.1700

Data Index: 1068314 (C4 (Web Text))

Max Activating Token Index: 845

Text #9

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6181. Min Act: -0.1700

Data Index: 384973 (C4 (Web Text))

Max Activating Token Index: 476

Text #10

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6025. Min Act: -0.1700

Data Index: 1124039 (C4 (Web Text))

Max Activating Token Index: 850

Text #11

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6338. Min Act: -0.1700

Data Index: 371713 (C4 (Web Text))

Max Activating Token Index: 862

Text #12

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6181. Min Act: -0.1700

Data Index: 1062267 (C4 (Web Text))

Max Activating Token Index: 920

Text #13

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.5868. Min Act: -0.1700

Data Index: 576578 (C4 (Web Text))

Max Activating Token Index: 648

Text #14

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6025. Min Act: -0.1700

Data Index: 394130 (C4 (Web Text))

Max Activating Token Index: 501

Text #15

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.5868. Min Act: -0.1700

Data Index: 74184 (C4 (Web Text))

Max Activating Token Index: 584

Text #16

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6025. Min Act: -0.1700

Data Index: 780457 (C4 (Web Text))

Max Activating Token Index: 935

Text #17

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.6181. Min Act: -0.1700

Data Index: 363757 (C4 (Web Text))

Max Activating Token Index: 631

Text #18

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.5711. Min Act: -0.1700

Data Index: 1198221 (C4 (Web Text))

Max Activating Token Index: 251

Text #19

Max Range: 3.8059. Min Range: -3.8059

Max Act: 3.5555. Min Act: -0.1700

Data Index: 105544 (C4 (Web Text))

Max Activating Token Index: 697