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

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

Neuron 7 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: 5.4034. Min Range: -5.4034

Max Act: 5.4034. Min Act: -0.1700

Data Index: 938722 (C4 (Web Text))

Max Activating Token Index: 293

Text #1

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.2472. Min Act: -0.1700

Data Index: 482304 (C4 (Web Text))

Max Activating Token Index: 402

Text #2

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.1847. Min Act: -0.1700

Data Index: 877812 (C4 (Web Text))

Max Activating Token Index: 998

Text #3

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.1847. Min Act: -0.1700

Data Index: 315371 (C4 (Web Text))

Max Activating Token Index: 472

Text #4

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.1847. Min Act: -0.1700

Data Index: 281334 (C4 (Web Text))

Max Activating Token Index: 31

Text #5

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.1847. Min Act: -0.1700

Data Index: 642651 (C4 (Web Text))

Max Activating Token Index: 558

Text #6

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.2159. Min Act: -0.1700

Data Index: 1251640 (C4 (Web Text))

Max Activating Token Index: 198

Text #7

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.1534. Min Act: -0.1700

Data Index: 667176 (C4 (Web Text))

Max Activating Token Index: 372

Text #8

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.0909. Min Act: -0.1700

Data Index: 704061 (C4 (Web Text))

Max Activating Token Index: 551

Text #9

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.0909. Min Act: -0.1700

Data Index: 628460 (C4 (Web Text))

Max Activating Token Index: 927

Text #10

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.0284. Min Act: -0.1700

Data Index: 1130118 (C4 (Web Text))

Max Activating Token Index: 507

Text #11

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.0597. Min Act: -0.1700

Data Index: 125093 (C4 (Web Text))

Max Activating Token Index: 999

Text #12

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9972. Min Act: -0.1700

Data Index: 62196 (C4 (Web Text))

Max Activating Token Index: 778

Text #13

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9659. Min Act: -0.1700

Data Index: 1057152 (C4 (Web Text))

Max Activating Token Index: 668

Text #14

Max Range: 5.4034. Min Range: -5.4034

Max Act: 5.0284. Min Act: -0.1700

Data Index: 241922 (C4 (Web Text))

Max Activating Token Index: 181

Text #15

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9972. Min Act: -0.1700

Data Index: 311356 (C4 (Web Text))

Max Activating Token Index: 362

Text #16

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9659. Min Act: -0.1700

Data Index: 1205750 (C4 (Web Text))

Max Activating Token Index: 421

Text #17

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9659. Min Act: -0.1700

Data Index: 423777 (C4 (Web Text))

Max Activating Token Index: 831

Text #18

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9659. Min Act: -0.1700

Data Index: 1113328 (C4 (Web Text))

Max Activating Token Index: 470

Text #19

Max Range: 5.4034. Min Range: -5.4034

Max Act: 4.9659. Min Act: -0.1700

Data Index: 89742 (C4 (Web Text))

Max Activating Token Index: 199