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

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

Neuron 512 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: 5.1502. Min Range: -5.1502

Max Act: 5.1502. Min Act: -0.1700

Data Index: 1324664 (C4 (Web Text))

Max Activating Token Index: 303

Text #1

Max Range: 5.1502. Min Range: -5.1502

Max Act: 5.0877. Min Act: -0.1700

Data Index: 409963 (C4 (Web Text))

Max Activating Token Index: 693

Text #2

Max Range: 5.1502. Min Range: -5.1502

Max Act: 5.0564. Min Act: -0.1700

Data Index: 780173 (C4 (Web Text))

Max Activating Token Index: 868

Text #3

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9314. Min Act: -0.1700

Data Index: 859420 (C4 (Web Text))

Max Activating Token Index: 583

Text #4

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9314. Min Act: -0.1700

Data Index: 531362 (C4 (Web Text))

Max Activating Token Index: 825

Text #5

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9002. Min Act: -0.1700

Data Index: 1149020 (C4 (Web Text))

Max Activating Token Index: 956

Text #6

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9314. Min Act: -0.1700

Data Index: 1301249 (C4 (Web Text))

Max Activating Token Index: 640

Text #7

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9002. Min Act: -0.1700

Data Index: 359888 (C4 (Web Text))

Max Activating Token Index: 306

Text #8

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.9314. Min Act: -0.1700

Data Index: 902471 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8377. Min Act: -0.1700

Data Index: 1344593 (C4 (Web Text))

Max Activating Token Index: 589

Text #10

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8064. Min Act: -0.1700

Data Index: 783908 (C4 (Web Text))

Max Activating Token Index: 527

Text #11

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8689. Min Act: -0.1700

Data Index: 617805 (C4 (Web Text))

Max Activating Token Index: 999

Text #12

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8689. Min Act: -0.1700

Data Index: 675708 (C4 (Web Text))

Max Activating Token Index: 902

Text #13

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8689. Min Act: -0.1700

Data Index: 1192512 (C4 (Web Text))

Max Activating Token Index: 376

Text #14

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.8064. Min Act: -0.1700

Data Index: 880746 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.7752. Min Act: -0.1700

Data Index: 376410 (C4 (Web Text))

Max Activating Token Index: 837

Text #16

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.7752. Min Act: -0.1700

Data Index: 19441 (C4 (Web Text))

Max Activating Token Index: 534

Text #17

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.7439. Min Act: -0.1700

Data Index: 793808 (C4 (Web Text))

Max Activating Token Index: 395

Text #18

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.7439. Min Act: -0.1700

Data Index: 1010050 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 5.1502. Min Range: -5.1502

Max Act: 4.7439. Min Act: -0.1700

Data Index: 622065 (C4 (Web Text))

Max Activating Token Index: 70