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

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

Neuron 1900 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: 5.1866. Min Range: -5.1866

Max Act: 5.1866. Min Act: -0.1700

Data Index: 481304 (C4 (Web Text))

Max Activating Token Index: 670

Text #1

Max Range: 5.1866. Min Range: -5.1866

Max Act: 5.1241. Min Act: -0.1700

Data Index: 635681 (C4 (Web Text))

Max Activating Token Index: 936

Text #2

Max Range: 5.1866. Min Range: -5.1866

Max Act: 5.1241. Min Act: -0.1700

Data Index: 99967 (C4 (Web Text))

Max Activating Token Index: 785

Text #3

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.9678. Min Act: -0.1700

Data Index: 400730 (C4 (Web Text))

Max Activating Token Index: 535

Text #4

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.9366. Min Act: -0.1700

Data Index: 545667 (C4 (Web Text))

Max Activating Token Index: 748

Text #5

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.9053. Min Act: -0.1700

Data Index: 478801 (C4 (Web Text))

Max Activating Token Index: 69

Text #6

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.9053. Min Act: -0.1700

Data Index: 220283 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.9053. Min Act: -0.1700

Data Index: 638494 (C4 (Web Text))

Max Activating Token Index: 66

Text #8

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8741. Min Act: -0.1700

Data Index: 1096070 (C4 (Web Text))

Max Activating Token Index: 216

Text #9

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8741. Min Act: -0.1700

Data Index: 734045 (C4 (Web Text))

Max Activating Token Index: 442

Text #10

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8741. Min Act: -0.1700

Data Index: 825354 (C4 (Web Text))

Max Activating Token Index: 618

Text #11

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8741. Min Act: -0.1700

Data Index: 1350010 (C4 (Web Text))

Max Activating Token Index: 233

Text #12

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8428. Min Act: -0.1700

Data Index: 1099941 (C4 (Web Text))

Max Activating Token Index: 240

Text #13

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8428. Min Act: -0.1700

Data Index: 422006 (C4 (Web Text))

Max Activating Token Index: 558

Text #14

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8741. Min Act: -0.1700

Data Index: 545671 (C4 (Web Text))

Max Activating Token Index: 783

Text #15

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8116. Min Act: -0.1700

Data Index: 858767 (C4 (Web Text))

Max Activating Token Index: 547

Text #16

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8116. Min Act: -0.1700

Data Index: 94230 (C4 (Web Text))

Max Activating Token Index: 955

Text #17

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8428. Min Act: -0.1700

Data Index: 1025653 (C4 (Web Text))

Max Activating Token Index: 398

Text #18

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.7803. Min Act: -0.1700

Data Index: 880049 (C4 (Web Text))

Max Activating Token Index: 90

Text #19

Max Range: 5.1866. Min Range: -5.1866

Max Act: 4.8116. Min Act: -0.1700

Data Index: 179633 (C4 (Web Text))

Max Activating Token Index: 440