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

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

Neuron 1252 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.3943. Min Act: -0.1700

Data Index: 1388906 (Python Code)

Max Activating Token Index: 11

Text #1

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.3472. Min Act: -0.1700

Data Index: 1639426 (Python Code)

Max Activating Token Index: 12

Text #2

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2844. Min Act: -0.1700

Data Index: 1404072 (Python Code)

Max Activating Token Index: 21

Text #3

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2687. Min Act: -0.1700

Data Index: 1491260 (Python Code)

Max Activating Token Index: 4

Text #4

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2687. Min Act: -0.1700

Data Index: 1420544 (Python Code)

Max Activating Token Index: 15

Text #5

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2373. Min Act: -0.1700

Data Index: 1363447 (Python Code)

Max Activating Token Index: 7

Text #6

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2215. Min Act: -0.1700

Data Index: 1408782 (Python Code)

Max Activating Token Index: 11

Text #7

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2058. Min Act: -0.1700

Data Index: 1542300 (Python Code)

Max Activating Token Index: 10

Text #8

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1901. Min Act: -0.1700

Data Index: 1576148 (Python Code)

Max Activating Token Index: 424

Text #9

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.2058. Min Act: -0.1700

Data Index: 1514675 (Python Code)

Max Activating Token Index: 9

Text #10

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1743. Min Act: -0.1700

Data Index: 1374558 (Python Code)

Max Activating Token Index: 13

Text #11

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1743. Min Act: -0.1700

Data Index: 1364562 (Python Code)

Max Activating Token Index: 28

Text #12

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1901. Min Act: -0.1700

Data Index: 1404663 (Python Code)

Max Activating Token Index: 6

Text #13

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1586. Min Act: -0.1700

Data Index: 1587302 (Python Code)

Max Activating Token Index: 191

Text #14

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1586. Min Act: -0.1700

Data Index: 1468926 (Python Code)

Max Activating Token Index: 9

Text #15

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1743. Min Act: -0.1700

Data Index: 1557474 (Python Code)

Max Activating Token Index: 730

Text #16

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1586. Min Act: -0.1700

Data Index: 1572991 (Python Code)

Max Activating Token Index: 427

Text #17

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1743. Min Act: -0.1700

Data Index: 1484109 (Python Code)

Max Activating Token Index: 5

Text #18

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1901. Min Act: -0.1700

Data Index: 1533505 (Python Code)

Max Activating Token Index: 22

Text #19

Max Range: 3.3943. Min Range: -3.3943

Max Act: 3.1743. Min Act: -0.1700

Data Index: 1430456 (Python Code)

Max Activating Token Index: 15