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

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

Neuron 300 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: 3.7535. Min Range: -3.7535

Max Act: 3.7535. Min Act: -0.1700

Data Index: 1069149 (C4 (Web Text))

Max Activating Token Index: 602

Text #1

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.7065. Min Act: -0.1700

Data Index: 1038911 (C4 (Web Text))

Max Activating Token Index: 380

Text #2

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6909. Min Act: -0.1700

Data Index: 154622 (C4 (Web Text))

Max Activating Token Index: 607

Text #3

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6596. Min Act: -0.1700

Data Index: 610200 (C4 (Web Text))

Max Activating Token Index: 849

Text #4

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6909. Min Act: -0.1700

Data Index: 324112 (C4 (Web Text))

Max Activating Token Index: 201

Text #5

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6596. Min Act: -0.1700

Data Index: 712464 (C4 (Web Text))

Max Activating Token Index: 461

Text #6

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6439. Min Act: -0.1700

Data Index: 879141 (C4 (Web Text))

Max Activating Token Index: 83

Text #7

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6909. Min Act: -0.1700

Data Index: 159163 (C4 (Web Text))

Max Activating Token Index: 867

Text #8

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6909. Min Act: -0.1700

Data Index: 502198 (C4 (Web Text))

Max Activating Token Index: 448

Text #9

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6439. Min Act: -0.1700

Data Index: 1090983 (C4 (Web Text))

Max Activating Token Index: 391

Text #10

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6596. Min Act: -0.1700

Data Index: 981626 (C4 (Web Text))

Max Activating Token Index: 654

Text #11

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6283. Min Act: -0.1700

Data Index: 952628 (C4 (Web Text))

Max Activating Token Index: 830

Text #12

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6283. Min Act: -0.1700

Data Index: 397185 (C4 (Web Text))

Max Activating Token Index: 119

Text #13

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6439. Min Act: -0.1700

Data Index: 819871 (C4 (Web Text))

Max Activating Token Index: 223

Text #14

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6439. Min Act: -0.1700

Data Index: 1145900 (C4 (Web Text))

Max Activating Token Index: 98

Text #15

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6283. Min Act: -0.1700

Data Index: 253971 (C4 (Web Text))

Max Activating Token Index: 871

Text #16

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6439. Min Act: -0.1700

Data Index: 1326366 (C4 (Web Text))

Max Activating Token Index: 146

Text #17

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6283. Min Act: -0.1700

Data Index: 460266 (C4 (Web Text))

Max Activating Token Index: 777

Text #18

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6126. Min Act: -0.1700

Data Index: 876732 (C4 (Web Text))

Max Activating Token Index: 318

Text #19

Max Range: 3.7535. Min Range: -3.7535

Max Act: 3.6283. Min Act: -0.1700

Data Index: 99096 (C4 (Web Text))

Max Activating Token Index: 526