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

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

Neuron 1609 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: 3.3283. Min Range: -3.3283

Max Act: 3.3283. Min Act: -0.1700

Data Index: 670 (C4 (Web Text))

Max Activating Token Index: 480

Text #1

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2812. Min Act: -0.1700

Data Index: 1001098 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2812. Min Act: -0.1700

Data Index: 257782 (C4 (Web Text))

Max Activating Token Index: 876

Text #3

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2655. Min Act: -0.1700

Data Index: 1338841 (C4 (Web Text))

Max Activating Token Index: 289

Text #4

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2183. Min Act: -0.1700

Data Index: 895600 (C4 (Web Text))

Max Activating Token Index: 872

Text #5

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2026. Min Act: -0.1700

Data Index: 880079 (C4 (Web Text))

Max Activating Token Index: 283

Text #6

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2340. Min Act: -0.1700

Data Index: 1291617 (C4 (Web Text))

Max Activating Token Index: 489

Text #7

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.2026. Min Act: -0.1700

Data Index: 948443 (C4 (Web Text))

Max Activating Token Index: 1019

Text #8

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1554. Min Act: -0.1700

Data Index: 224864 (C4 (Web Text))

Max Activating Token Index: 197

Text #9

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1711. Min Act: -0.1700

Data Index: 856409 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1396. Min Act: -0.1700

Data Index: 409378 (C4 (Web Text))

Max Activating Token Index: 781

Text #11

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1396. Min Act: -0.1700

Data Index: 1031196 (C4 (Web Text))

Max Activating Token Index: 540

Text #12

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1554. Min Act: -0.1700

Data Index: 844159 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1711. Min Act: -0.1700

Data Index: 1040100 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1396. Min Act: -0.1700

Data Index: 200305 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1396. Min Act: -0.1700

Data Index: 882614 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1081. Min Act: -0.1700

Data Index: 263197 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1239. Min Act: -0.1700

Data Index: 738637 (C4 (Web Text))

Max Activating Token Index: 678

Text #18

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1081. Min Act: -0.1700

Data Index: 653288 (C4 (Web Text))

Max Activating Token Index: 473

Text #19

Max Range: 3.3283. Min Range: -3.3283

Max Act: 3.1081. Min Act: -0.1700

Data Index: 617760 (C4 (Web Text))

Max Activating Token Index: 173