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

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

Neuron 1509 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1845. Min Range: -5.1845

Max Act: 5.1845. Min Act: -0.1700

Data Index: 1067121 (C4 (Web Text))

Max Activating Token Index: 161

Text #1

Max Range: 5.1845. Min Range: -5.1845

Max Act: 5.1532. Min Act: -0.1700

Data Index: 1196272 (C4 (Web Text))

Max Activating Token Index: 337

Text #2

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9970. Min Act: -0.1700

Data Index: 759339 (C4 (Web Text))

Max Activating Token Index: 415

Text #3

Max Range: 5.1845. Min Range: -5.1845

Max Act: 5.0282. Min Act: -0.1700

Data Index: 830478 (C4 (Web Text))

Max Activating Token Index: 757

Text #4

Max Range: 5.1845. Min Range: -5.1845

Max Act: 5.0126. Min Act: -0.1700

Data Index: 135594 (C4 (Web Text))

Max Activating Token Index: 546

Text #5

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9345. Min Act: -0.1700

Data Index: 410162 (C4 (Web Text))

Max Activating Token Index: 950

Text #6

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9501. Min Act: -0.1700

Data Index: 444579 (C4 (Web Text))

Max Activating Token Index: 542

Text #7

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9501. Min Act: -0.1700

Data Index: 424717 (C4 (Web Text))

Max Activating Token Index: 717

Text #8

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9345. Min Act: -0.1700

Data Index: 330326 (C4 (Web Text))

Max Activating Token Index: 905

Text #9

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9345. Min Act: -0.1700

Data Index: 1224804 (C4 (Web Text))

Max Activating Token Index: 161

Text #10

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8876. Min Act: -0.1700

Data Index: 527891 (C4 (Web Text))

Max Activating Token Index: 720

Text #11

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9032. Min Act: -0.1700

Data Index: 1246980 (C4 (Web Text))

Max Activating Token Index: 196

Text #12

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9032. Min Act: -0.1700

Data Index: 596326 (C4 (Web Text))

Max Activating Token Index: 594

Text #13

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8564. Min Act: -0.1700

Data Index: 168542 (C4 (Web Text))

Max Activating Token Index: 968

Text #14

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.9189. Min Act: -0.1700

Data Index: 163363 (C4 (Web Text))

Max Activating Token Index: 429

Text #15

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8876. Min Act: -0.1700

Data Index: 1050189 (C4 (Web Text))

Max Activating Token Index: 361

Text #16

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8876. Min Act: -0.1700

Data Index: 710856 (C4 (Web Text))

Max Activating Token Index: 1020

Text #17

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8564. Min Act: -0.1700

Data Index: 752426 (C4 (Web Text))

Max Activating Token Index: 800

Text #18

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8564. Min Act: -0.1700

Data Index: 199423 (C4 (Web Text))

Max Activating Token Index: 500

Text #19

Max Range: 5.1845. Min Range: -5.1845

Max Act: 4.8720. Min Act: -0.1700

Data Index: 1464338 (Python Code)

Max Activating Token Index: 220