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

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

Neuron 1695 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: 7.3027. Min Range: -7.3027

Max Act: 7.3027. Min Act: -0.1700

Data Index: 1538253 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 7.3027. Min Range: -7.3027

Max Act: 7.1777. Min Act: -0.1700

Data Index: 1549263 (Python Code)

Max Activating Token Index: 876

Text #2

Max Range: 7.3027. Min Range: -7.3027

Max Act: 7.1152. Min Act: -0.1700

Data Index: 1591925 (Python Code)

Max Activating Token Index: 576

Text #3

Max Range: 7.3027. Min Range: -7.3027

Max Act: 7.1152. Min Act: -0.1700

Data Index: 1501754 (Python Code)

Max Activating Token Index: 587

Text #4

Max Range: 7.3027. Min Range: -7.3027

Max Act: 7.0527. Min Act: -0.1700

Data Index: 1376310 (Python Code)

Max Activating Token Index: 178

Text #5

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.9589. Min Act: -0.1700

Data Index: 1481831 (Python Code)

Max Activating Token Index: 906

Text #6

Max Range: 7.3027. Min Range: -7.3027

Max Act: 7.0527. Min Act: -0.1700

Data Index: 1646076 (Python Code)

Max Activating Token Index: 167

Text #7

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.8027. Min Act: -0.1700

Data Index: 1497318 (Python Code)

Max Activating Token Index: 474

Text #8

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.9902. Min Act: -0.1700

Data Index: 1414628 (Python Code)

Max Activating Token Index: 55

Text #9

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.8652. Min Act: -0.1700

Data Index: 1491896 (Python Code)

Max Activating Token Index: 88

Text #10

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.8964. Min Act: -0.1700

Data Index: 1582910 (Python Code)

Max Activating Token Index: 631

Text #11

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.8027. Min Act: -0.1700

Data Index: 1614787 (Python Code)

Max Activating Token Index: 976

Text #12

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.7402. Min Act: -0.1700

Data Index: 1393814 (Python Code)

Max Activating Token Index: 841

Text #13

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.7402. Min Act: -0.1700

Data Index: 1530918 (Python Code)

Max Activating Token Index: 442

Text #14

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.7402. Min Act: -0.1700

Data Index: 1560367 (Python Code)

Max Activating Token Index: 374

Text #15

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.4902. Min Act: -0.1700

Data Index: 1650392 (Python Code)

Max Activating Token Index: 609

Text #16

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.7714. Min Act: -0.1700

Data Index: 1409863 (Python Code)

Max Activating Token Index: 724

Text #17

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.6152. Min Act: -0.1700

Data Index: 1378747 (Python Code)

Max Activating Token Index: 191

Text #18

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.6152. Min Act: -0.1700

Data Index: 1521024 (Python Code)

Max Activating Token Index: 262

Text #19

Max Range: 7.3027. Min Range: -7.3027

Max Act: 6.6464. Min Act: -0.1700

Data Index: 1544491 (Python Code)

Max Activating Token Index: 976