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

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

Neuron 104 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: 3.3878. Min Range: -3.3878

Max Act: 3.3878. Min Act: -0.1700

Data Index: 1036939 (C4 (Web Text))

Max Activating Token Index: 1004

Text #1

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3878. Min Act: -0.1700

Data Index: 895326 (C4 (Web Text))

Max Activating Token Index: 593

Text #2

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3564. Min Act: -0.1700

Data Index: 1165154 (C4 (Web Text))

Max Activating Token Index: 208

Text #3

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3721. Min Act: -0.1700

Data Index: 812767 (C4 (Web Text))

Max Activating Token Index: 449

Text #4

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3407. Min Act: -0.1700

Data Index: 1017746 (C4 (Web Text))

Max Activating Token Index: 863

Text #5

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3407. Min Act: -0.1700

Data Index: 1177092 (C4 (Web Text))

Max Activating Token Index: 564

Text #6

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3721. Min Act: -0.1700

Data Index: 170115 (C4 (Web Text))

Max Activating Token Index: 412

Text #7

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3250. Min Act: -0.1700

Data Index: 179679 (C4 (Web Text))

Max Activating Token Index: 526

Text #8

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3250. Min Act: -0.1700

Data Index: 456263 (C4 (Web Text))

Max Activating Token Index: 513

Text #9

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 294821 (C4 (Web Text))

Max Activating Token Index: 912

Text #10

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 1324232 (C4 (Web Text))

Max Activating Token Index: 453

Text #11

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 381717 (C4 (Web Text))

Max Activating Token Index: 547

Text #12

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 464223 (C4 (Web Text))

Max Activating Token Index: 567

Text #13

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.2936. Min Act: -0.1700

Data Index: 462644 (C4 (Web Text))

Max Activating Token Index: 165

Text #14

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 1222298 (C4 (Web Text))

Max Activating Token Index: 749

Text #15

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.3093. Min Act: -0.1700

Data Index: 516754 (C4 (Web Text))

Max Activating Token Index: 682

Text #16

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.2779. Min Act: -0.1700

Data Index: 400300 (C4 (Web Text))

Max Activating Token Index: 781

Text #17

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.2779. Min Act: -0.1700

Data Index: 997538 (C4 (Web Text))

Max Activating Token Index: 164

Text #18

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.2936. Min Act: -0.1700

Data Index: 508842 (C4 (Web Text))

Max Activating Token Index: 679

Text #19

Max Range: 3.3878. Min Range: -3.3878

Max Act: 3.2779. Min Act: -0.1700

Data Index: 294293 (C4 (Web Text))

Max Activating Token Index: 66