Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 890 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.4474. Min Act: -0.1700

Data Index: 685496 (C4 (Web Text))

Max Activating Token Index: 332

Text #1

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.4161. Min Act: -0.1700

Data Index: 1491081 (Python Code)

Max Activating Token Index: 568

Text #2

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.4161. Min Act: -0.1700

Data Index: 1616859 (Python Code)

Max Activating Token Index: 670

Text #3

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3847. Min Act: -0.1700

Data Index: 600437 (C4 (Web Text))

Max Activating Token Index: 817

Text #4

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3533. Min Act: -0.1700

Data Index: 1377359 (Python Code)

Max Activating Token Index: 378

Text #5

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3533. Min Act: -0.1681

Data Index: 897821 (C4 (Web Text))

Max Activating Token Index: 753

Text #6

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3533. Min Act: -0.1700

Data Index: 719788 (C4 (Web Text))

Max Activating Token Index: 776

Text #7

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3219. Min Act: -0.1403

Data Index: 587241 (C4 (Web Text))

Max Activating Token Index: 593

Text #8

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1680

Data Index: 349978 (C4 (Web Text))

Max Activating Token Index: 272

Text #9

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 1534092 (Python Code)

Max Activating Token Index: 235

Text #10

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 1392169 (Python Code)

Max Activating Token Index: 981

Text #11

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 1646811 (Python Code)

Max Activating Token Index: 688

Text #12

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 1474009 (Python Code)

Max Activating Token Index: 985

Text #13

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.3219. Min Act: -0.1700

Data Index: 1581705 (Python Code)

Max Activating Token Index: 783

Text #14

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 174789 (C4 (Web Text))

Max Activating Token Index: 240

Text #15

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2905. Min Act: -0.1700

Data Index: 326969 (C4 (Web Text))

Max Activating Token Index: 646

Text #16

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2591. Min Act: -0.1700

Data Index: 1567291 (Python Code)

Max Activating Token Index: 568

Text #17

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2591. Min Act: -0.1700

Data Index: 1603923 (Python Code)

Max Activating Token Index: 185

Text #18

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2434. Min Act: -0.1700

Data Index: 691370 (C4 (Web Text))

Max Activating Token Index: 490

Text #19

Max Range: 3.4474. Min Range: -3.4474

Max Act: 3.2434. Min Act: -0.1700

Data Index: 762521 (C4 (Web Text))

Max Activating Token Index: 558