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

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

Neuron 1766 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2560. Min Range: -5.2560

Max Act: 5.2560. Min Act: -0.1700

Data Index: 1582060 (Python Code)

Max Activating Token Index: 817

Text #1

Max Range: 5.2560. Min Range: -5.2560

Max Act: 5.2247. Min Act: -0.1700

Data Index: 1501340 (Python Code)

Max Activating Token Index: 658

Text #2

Max Range: 5.2560. Min Range: -5.2560

Max Act: 5.1622. Min Act: -0.1700

Data Index: 1550447 (Python Code)

Max Activating Token Index: 739

Text #3

Max Range: 5.2560. Min Range: -5.2560

Max Act: 5.0060. Min Act: -0.1700

Data Index: 35195 (C4 (Web Text))

Max Activating Token Index: 777

Text #4

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.9747. Min Act: -0.1700

Data Index: 1526684 (Python Code)

Max Activating Token Index: 213

Text #5

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.9278. Min Act: -0.1700

Data Index: 1377280 (Python Code)

Max Activating Token Index: 924

Text #6

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.8966. Min Act: -0.1700

Data Index: 1480919 (Python Code)

Max Activating Token Index: 933

Text #7

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.9122. Min Act: -0.1700

Data Index: 1382667 (Python Code)

Max Activating Token Index: 1002

Text #8

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.8810. Min Act: -0.1700

Data Index: 1400114 (Python Code)

Max Activating Token Index: 841

Text #9

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.8653. Min Act: -0.1700

Data Index: 1369225 (Python Code)

Max Activating Token Index: 912

Text #10

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.8341. Min Act: -0.1700

Data Index: 1561927 (Python Code)

Max Activating Token Index: 558

Text #11

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.7716. Min Act: -0.1700

Data Index: 1373111 (Python Code)

Max Activating Token Index: 875

Text #12

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.8028. Min Act: -0.1700

Data Index: 591056 (C4 (Web Text))

Max Activating Token Index: 463

Text #13

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.7716. Min Act: -0.1700

Data Index: 1575178 (Python Code)

Max Activating Token Index: 899

Text #14

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.7247. Min Act: -0.1700

Data Index: 1522771 (Python Code)

Max Activating Token Index: 758

Text #15

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.6935. Min Act: -0.1700

Data Index: 1481374 (Python Code)

Max Activating Token Index: 421

Text #16

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.7091. Min Act: -0.1700

Data Index: 1399881 (Python Code)

Max Activating Token Index: 496

Text #17

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.6622. Min Act: -0.1700

Data Index: 1600169 (Python Code)

Max Activating Token Index: 853

Text #18

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.6778. Min Act: -0.1700

Data Index: 1499162 (Python Code)

Max Activating Token Index: 827

Text #19

Max Range: 5.2560. Min Range: -5.2560

Max Act: 4.5997. Min Act: -0.1700

Data Index: 1584361 (Python Code)

Max Activating Token Index: 1016