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

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

Neuron 1298 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: 3.2986. Min Range: -3.2986

Max Act: 3.2986. Min Act: -0.1700

Data Index: 1615293 (Python Code)

Max Activating Token Index: 63

Text #1

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.2672. Min Act: -0.1700

Data Index: 1617907 (Python Code)

Max Activating Token Index: 288

Text #2

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.2200. Min Act: -0.1700

Data Index: 1626719 (Python Code)

Max Activating Token Index: 25

Text #3

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1886. Min Act: -0.1700

Data Index: 1486610 (Python Code)

Max Activating Token Index: 203

Text #4

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1886. Min Act: -0.1700

Data Index: 1606799 (Python Code)

Max Activating Token Index: 65

Text #5

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.2043. Min Act: -0.1700

Data Index: 1656439 (Python Code)

Max Activating Token Index: 250

Text #6

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1728. Min Act: -0.1700

Data Index: 1410593 (Python Code)

Max Activating Token Index: 839

Text #7

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1571. Min Act: -0.1700

Data Index: 1095199 (C4 (Web Text))

Max Activating Token Index: 3

Text #8

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1886. Min Act: -0.1700

Data Index: 1458729 (Python Code)

Max Activating Token Index: 723

Text #9

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1256. Min Act: -0.1700

Data Index: 936922 (C4 (Web Text))

Max Activating Token Index: 53

Text #10

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1256. Min Act: -0.1700

Data Index: 1136588 (C4 (Web Text))

Max Activating Token Index: 70

Text #11

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1098. Min Act: -0.1700

Data Index: 1429623 (Python Code)

Max Activating Token Index: 689

Text #12

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1256. Min Act: -0.1700

Data Index: 94407 (C4 (Web Text))

Max Activating Token Index: 612

Text #13

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.1098. Min Act: -0.1700

Data Index: 150792 (C4 (Web Text))

Max Activating Token Index: 471

Text #14

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0940. Min Act: -0.1700

Data Index: 1620208 (Python Code)

Max Activating Token Index: 990

Text #15

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0783. Min Act: -0.1700

Data Index: 1586092 (Python Code)

Max Activating Token Index: 586

Text #16

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0783. Min Act: -0.1700

Data Index: 1485082 (Python Code)

Max Activating Token Index: 271

Text #17

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0940. Min Act: -0.1700

Data Index: 352714 (C4 (Web Text))

Max Activating Token Index: 492

Text #18

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0783. Min Act: -0.1700

Data Index: 692631 (C4 (Web Text))

Max Activating Token Index: 515

Text #19

Max Range: 3.2986. Min Range: -3.2986

Max Act: 3.0625. Min Act: -0.1700

Data Index: 1602450 (Python Code)

Max Activating Token Index: 900