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

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

Neuron 711 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: 3.5396. Min Range: -3.5396

Max Act: 3.5396. Min Act: -0.1700

Data Index: 1620265 (Python Code)

Max Activating Token Index: 66

Text #1

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.3514. Min Act: -0.1700

Data Index: 48131 (C4 (Web Text))

Max Activating Token Index: 614

Text #2

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.4142. Min Act: -0.1700

Data Index: 1502559 (Python Code)

Max Activating Token Index: 189

Text #3

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.4142. Min Act: -0.1700

Data Index: 1434689 (Python Code)

Max Activating Token Index: 41

Text #4

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.3514. Min Act: -0.1700

Data Index: 1447651 (Python Code)

Max Activating Token Index: 23

Text #5

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.3828. Min Act: -0.1700

Data Index: 1522385 (Python Code)

Max Activating Token Index: 46

Text #6

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.3200. Min Act: -0.1700

Data Index: 1596551 (Python Code)

Max Activating Token Index: 155

Text #7

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.3514. Min Act: -0.1700

Data Index: 1629115 (Python Code)

Max Activating Token Index: 339

Text #8

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2886. Min Act: -0.1700

Data Index: 1509713 (Python Code)

Max Activating Token Index: 18

Text #9

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2886. Min Act: -0.1700

Data Index: 1481174 (Python Code)

Max Activating Token Index: 920

Text #10

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2886. Min Act: -0.1700

Data Index: 1638907 (Python Code)

Max Activating Token Index: 732

Text #11

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2886. Min Act: -0.1700

Data Index: 1542588 (Python Code)

Max Activating Token Index: 717

Text #12

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2886. Min Act: -0.1700

Data Index: 1532832 (Python Code)

Max Activating Token Index: 557

Text #13

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2258. Min Act: -0.1700

Data Index: 1557138 (Python Code)

Max Activating Token Index: 749

Text #14

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.1786. Min Act: -0.1700

Data Index: 1384772 (Python Code)

Max Activating Token Index: 277

Text #15

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2258. Min Act: -0.1700

Data Index: 1596218 (Python Code)

Max Activating Token Index: 672

Text #16

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2258. Min Act: -0.1700

Data Index: 1489828 (Python Code)

Max Activating Token Index: 393

Text #17

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2572. Min Act: -0.1700

Data Index: 1645864 (Python Code)

Max Activating Token Index: 436

Text #18

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2572. Min Act: -0.1700

Data Index: 1402472 (Python Code)

Max Activating Token Index: 49

Text #19

Max Range: 3.5396. Min Range: -3.5396

Max Act: 3.2572. Min Act: -0.1700

Data Index: 1431141 (Python Code)

Max Activating Token Index: 606