Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 2036 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.6572. Min Act: -0.1700

Data Index: 65340 (C4 (Web Text))

Max Activating Token Index: 855

Text #1

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.6259. Min Act: -0.1700

Data Index: 1037417 (C4 (Web Text))

Max Activating Token Index: 961

Text #2

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.5790. Min Act: -0.1700

Data Index: 580548 (C4 (Web Text))

Max Activating Token Index: 348

Text #3

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.5477. Min Act: -0.1700

Data Index: 702065 (C4 (Web Text))

Max Activating Token Index: 589

Text #4

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.5477. Min Act: -0.1700

Data Index: 65604 (C4 (Web Text))

Max Activating Token Index: 721

Text #5

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.5320. Min Act: -0.1700

Data Index: 1201057 (C4 (Web Text))

Max Activating Token Index: 867

Text #6

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4850. Min Act: -0.1700

Data Index: 919601 (C4 (Web Text))

Max Activating Token Index: 45

Text #7

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4693. Min Act: -0.1700

Data Index: 93006 (C4 (Web Text))

Max Activating Token Index: 204

Text #8

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.5320. Min Act: -0.1700

Data Index: 166057 (C4 (Web Text))

Max Activating Token Index: 57

Text #9

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4693. Min Act: -0.1700

Data Index: 1181169 (C4 (Web Text))

Max Activating Token Index: 591

Text #10

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4693. Min Act: -0.1700

Data Index: 836752 (C4 (Web Text))

Max Activating Token Index: 551

Text #11

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4536. Min Act: -0.1700

Data Index: 124310 (C4 (Web Text))

Max Activating Token Index: 958

Text #12

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4693. Min Act: -0.1700

Data Index: 121596 (C4 (Web Text))

Max Activating Token Index: 561

Text #13

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4850. Min Act: -0.1700

Data Index: 1297673 (C4 (Web Text))

Max Activating Token Index: 681

Text #14

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4223. Min Act: -0.1700

Data Index: 1346756 (C4 (Web Text))

Max Activating Token Index: 727

Text #15

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4536. Min Act: -0.1700

Data Index: 237552 (C4 (Web Text))

Max Activating Token Index: 256

Text #16

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4223. Min Act: -0.1700

Data Index: 791706 (C4 (Web Text))

Max Activating Token Index: 713

Text #17

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4223. Min Act: -0.1700

Data Index: 339163 (C4 (Web Text))

Max Activating Token Index: 956

Text #18

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4380. Min Act: -0.1700

Data Index: 419819 (C4 (Web Text))

Max Activating Token Index: 500

Text #19

Max Range: 3.6572. Min Range: -3.6572

Max Act: 3.4066. Min Act: -0.1700

Data Index: 1310562 (C4 (Web Text))

Max Activating Token Index: 693