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

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

Neuron 1106 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: 7.1830. Min Range: -7.1830

Max Act: 7.1830. Min Act: -0.1700

Data Index: 1076374 (C4 (Web Text))

Max Activating Token Index: 921

Text #1

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.9643. Min Act: -0.1700

Data Index: 1413266 (Python Code)

Max Activating Token Index: 860

Text #2

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8393. Min Act: -0.1700

Data Index: 1090822 (C4 (Web Text))

Max Activating Token Index: 937

Text #3

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.9018. Min Act: -0.1700

Data Index: 1530564 (Python Code)

Max Activating Token Index: 398

Text #4

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8393. Min Act: -0.1700

Data Index: 1593051 (Python Code)

Max Activating Token Index: 86

Text #5

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8393. Min Act: -0.1700

Data Index: 1616948 (Python Code)

Max Activating Token Index: 132

Text #6

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7768. Min Act: -0.1700

Data Index: 1630577 (Python Code)

Max Activating Token Index: 231

Text #7

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8393. Min Act: -0.1700

Data Index: 1473753 (Python Code)

Max Activating Token Index: 439

Text #8

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8705. Min Act: -0.1700

Data Index: 1361304 (Python Code)

Max Activating Token Index: 311

Text #9

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7768. Min Act: -0.1700

Data Index: 1027529 (C4 (Web Text))

Max Activating Token Index: 773

Text #10

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7768. Min Act: -0.1700

Data Index: 1384720 (Python Code)

Max Activating Token Index: 307

Text #11

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7143. Min Act: -0.1700

Data Index: 1628399 (Python Code)

Max Activating Token Index: 349

Text #12

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7455. Min Act: -0.1700

Data Index: 1473049 (Python Code)

Max Activating Token Index: 414

Text #13

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.6830. Min Act: -0.1700

Data Index: 1379808 (Python Code)

Max Activating Token Index: 619

Text #14

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.7143. Min Act: -0.1700

Data Index: 1520846 (Python Code)

Max Activating Token Index: 697

Text #15

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.6830. Min Act: -0.1700

Data Index: 1516972 (Python Code)

Max Activating Token Index: 477

Text #16

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.6830. Min Act: -0.1700

Data Index: 1389046 (Python Code)

Max Activating Token Index: 381

Text #17

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.8080. Min Act: -0.1700

Data Index: 1442213 (Python Code)

Max Activating Token Index: 266

Text #18

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.6830. Min Act: -0.1700

Data Index: 1386719 (Python Code)

Max Activating Token Index: 501

Text #19

Max Range: 7.1830. Min Range: -7.1830

Max Act: 6.6518. Min Act: -0.1700

Data Index: 684777 (C4 (Web Text))

Max Activating Token Index: 698