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

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

Neuron 550 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: 5.3278. Min Range: -5.3278

Max Act: 5.3278. Min Act: -0.1700

Data Index: 302675 (C4 (Web Text))

Max Activating Token Index: 426

Text #1

Max Range: 5.3278. Min Range: -5.3278

Max Act: 5.2966. Min Act: -0.1700

Data Index: 746108 (C4 (Web Text))

Max Activating Token Index: 451

Text #2

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.9216. Min Act: -0.1700

Data Index: 133317 (C4 (Web Text))

Max Activating Token Index: 474

Text #3

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.8435. Min Act: -0.1700

Data Index: 1201181 (C4 (Web Text))

Max Activating Token Index: 685

Text #4

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.8591. Min Act: -0.1700

Data Index: 357419 (C4 (Web Text))

Max Activating Token Index: 339

Text #5

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.7810. Min Act: -0.1700

Data Index: 315132 (C4 (Web Text))

Max Activating Token Index: 661

Text #6

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.7966. Min Act: -0.1700

Data Index: 69357 (C4 (Web Text))

Max Activating Token Index: 156

Text #7

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.5935. Min Act: -0.1700

Data Index: 602113 (C4 (Web Text))

Max Activating Token Index: 393

Text #8

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6716. Min Act: -0.1700

Data Index: 1023243 (C4 (Web Text))

Max Activating Token Index: 939

Text #9

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.7028. Min Act: -0.1700

Data Index: 145642 (C4 (Web Text))

Max Activating Token Index: 273

Text #10

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6560. Min Act: -0.1700

Data Index: 694708 (C4 (Web Text))

Max Activating Token Index: 549

Text #11

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6403. Min Act: -0.1700

Data Index: 32195 (C4 (Web Text))

Max Activating Token Index: 231

Text #12

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6560. Min Act: -0.1700

Data Index: 1260729 (C4 (Web Text))

Max Activating Token Index: 473

Text #13

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6403. Min Act: -0.1700

Data Index: 559231 (C4 (Web Text))

Max Activating Token Index: 564

Text #14

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6091. Min Act: -0.1700

Data Index: 1154731 (C4 (Web Text))

Max Activating Token Index: 142

Text #15

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.5935. Min Act: -0.1700

Data Index: 708480 (C4 (Web Text))

Max Activating Token Index: 592

Text #16

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6091. Min Act: -0.1700

Data Index: 1214804 (C4 (Web Text))

Max Activating Token Index: 603

Text #17

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6872. Min Act: -0.1700

Data Index: 764287 (C4 (Web Text))

Max Activating Token Index: 297

Text #18

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6247. Min Act: -0.1700

Data Index: 901163 (C4 (Web Text))

Max Activating Token Index: 196

Text #19

Max Range: 5.3278. Min Range: -5.3278

Max Act: 4.6247. Min Act: -0.1700

Data Index: 529698 (C4 (Web Text))

Max Activating Token Index: 913