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

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

Neuron 128 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.1922. Min Act: -0.1659

Data Index: 499573 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.1297. Min Act: -0.1699

Data Index: 909886 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0672. Min Act: -0.1699

Data Index: 725424 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0516. Min Act: -0.1699

Data Index: 1355007 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0516. Min Act: -0.1700

Data Index: 204385 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0359. Min Act: -0.1698

Data Index: 1307825 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0359. Min Act: -0.1699

Data Index: 895383 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 4.1922. Min Range: -4.1922

Max Act: 4.0203. Min Act: -0.1700

Data Index: 183785 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9734. Min Act: -0.1700

Data Index: 576144 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9734. Min Act: -0.1699

Data Index: 1140429 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9421. Min Act: -0.1699

Data Index: 939182 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9578. Min Act: -0.1698

Data Index: 298513 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9421. Min Act: -0.1699

Data Index: 651749 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9578. Min Act: -0.1699

Data Index: 1044840 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9421. Min Act: -0.1698

Data Index: 660690 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9109. Min Act: -0.1562

Data Index: 329048 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9109. Min Act: -0.1699

Data Index: 185645 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9109. Min Act: -0.1700

Data Index: 772436 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.9265. Min Act: -0.1700

Data Index: 491379 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 4.1922. Min Range: -4.1922

Max Act: 3.8952. Min Act: -0.1698

Data Index: 1312799 (C4 (Web Text))

Max Activating Token Index: 1023