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

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

Neuron 555 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: 5.2197. Min Range: -5.2197

Max Act: 5.2197. Min Act: -0.1700

Data Index: 436085 (C4 (Web Text))

Max Activating Token Index: 655

Text #1

Max Range: 5.2197. Min Range: -5.2197

Max Act: 5.0478. Min Act: -0.1700

Data Index: 1181896 (C4 (Web Text))

Max Activating Token Index: 619

Text #2

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.9540. Min Act: -0.1700

Data Index: 725677 (C4 (Web Text))

Max Activating Token Index: 150

Text #3

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8915. Min Act: -0.1700

Data Index: 1193294 (C4 (Web Text))

Max Activating Token Index: 164

Text #4

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8759. Min Act: -0.1700

Data Index: 658443 (C4 (Web Text))

Max Activating Token Index: 618

Text #5

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8603. Min Act: -0.1700

Data Index: 180712 (C4 (Web Text))

Max Activating Token Index: 871

Text #6

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8759. Min Act: -0.1700

Data Index: 1101167 (C4 (Web Text))

Max Activating Token Index: 52

Text #7

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8603. Min Act: -0.1700

Data Index: 374075 (C4 (Web Text))

Max Activating Token Index: 1003

Text #8

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8447. Min Act: -0.1700

Data Index: 181524 (C4 (Web Text))

Max Activating Token Index: 295

Text #9

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7978. Min Act: -0.1699

Data Index: 609788 (C4 (Web Text))

Max Activating Token Index: 526

Text #10

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8290. Min Act: -0.1700

Data Index: 820488 (C4 (Web Text))

Max Activating Token Index: 926

Text #11

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7978. Min Act: -0.1700

Data Index: 1148459 (C4 (Web Text))

Max Activating Token Index: 540

Text #12

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8134. Min Act: -0.1700

Data Index: 695069 (C4 (Web Text))

Max Activating Token Index: 669

Text #13

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.8134. Min Act: -0.1700

Data Index: 1258666 (C4 (Web Text))

Max Activating Token Index: 182

Text #14

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7665. Min Act: -0.1700

Data Index: 40964 (C4 (Web Text))

Max Activating Token Index: 526

Text #15

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.5947. Min Act: -0.1700

Data Index: 1093961 (C4 (Web Text))

Max Activating Token Index: 762

Text #16

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7197. Min Act: -0.1700

Data Index: 1017623 (C4 (Web Text))

Max Activating Token Index: 536

Text #17

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7353. Min Act: -0.1700

Data Index: 747597 (C4 (Web Text))

Max Activating Token Index: 592

Text #18

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7197. Min Act: -0.1700

Data Index: 85075 (C4 (Web Text))

Max Activating Token Index: 198

Text #19

Max Range: 5.2197. Min Range: -5.2197

Max Act: 4.7197. Min Act: -0.1699

Data Index: 718079 (C4 (Web Text))

Max Activating Token Index: 508