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

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

Neuron 1839 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0169. Min Range: -7.0169

Max Act: 7.0169. Min Act: -0.1700

Data Index: 1644115 (Python Code)

Max Activating Token Index: 656

Text #1

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.9544. Min Act: -0.1700

Data Index: 1216952 (C4 (Web Text))

Max Activating Token Index: 1011

Text #2

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.6731. Min Act: -0.1700

Data Index: 1461050 (Python Code)

Max Activating Token Index: 104

Text #3

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.6106. Min Act: -0.1700

Data Index: 1526027 (Python Code)

Max Activating Token Index: 800

Text #4

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.6419. Min Act: -0.1700

Data Index: 1577048 (Python Code)

Max Activating Token Index: 754

Text #5

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.5169. Min Act: -0.1700

Data Index: 1585279 (Python Code)

Max Activating Token Index: 982

Text #6

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.4856. Min Act: -0.1700

Data Index: 1554635 (Python Code)

Max Activating Token Index: 1006

Text #7

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.4544. Min Act: -0.1700

Data Index: 1437683 (Python Code)

Max Activating Token Index: 969

Text #8

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.4544. Min Act: -0.1700

Data Index: 1468040 (Python Code)

Max Activating Token Index: 370

Text #9

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.3919. Min Act: -0.1700

Data Index: 1638229 (Python Code)

Max Activating Token Index: 513

Text #10

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.4231. Min Act: -0.1700

Data Index: 1484905 (Python Code)

Max Activating Token Index: 961

Text #11

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.3294. Min Act: -0.1700

Data Index: 1530438 (Python Code)

Max Activating Token Index: 404

Text #12

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.2981. Min Act: -0.1700

Data Index: 1538492 (Python Code)

Max Activating Token Index: 596

Text #13

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.3919. Min Act: -0.1700

Data Index: 305088 (C4 (Web Text))

Max Activating Token Index: 1016

Text #14

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.3606. Min Act: -0.1700

Data Index: 1492778 (Python Code)

Max Activating Token Index: 988

Text #15

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.1731. Min Act: -0.1700

Data Index: 1414104 (Python Code)

Max Activating Token Index: 432

Text #16

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.3294. Min Act: -0.1700

Data Index: 1525968 (Python Code)

Max Activating Token Index: 572

Text #17

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.2044. Min Act: -0.1700

Data Index: 1264213 (C4 (Web Text))

Max Activating Token Index: 590

Text #18

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.2981. Min Act: -0.1700

Data Index: 1531620 (Python Code)

Max Activating Token Index: 213

Text #19

Max Range: 7.0169. Min Range: -7.0169

Max Act: 6.2356. Min Act: -0.1700

Data Index: 1446227 (Python Code)

Max Activating Token Index: 893