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

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

Neuron 1041 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.7020. Min Act: -0.1700

Data Index: 1624617 (Python Code)

Max Activating Token Index: 284

Text #1

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.6394. Min Act: -0.1700

Data Index: 941500 (C4 (Web Text))

Max Activating Token Index: 108

Text #2

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.6081. Min Act: -0.1700

Data Index: 1200112 (C4 (Web Text))

Max Activating Token Index: 855

Text #3

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.6081. Min Act: -0.1700

Data Index: 250657 (C4 (Web Text))

Max Activating Token Index: 136

Text #4

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5455. Min Act: -0.1700

Data Index: 1255673 (C4 (Web Text))

Max Activating Token Index: 51

Text #5

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5768. Min Act: -0.1700

Data Index: 1361725 (Python Code)

Max Activating Token Index: 278

Text #6

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5141. Min Act: -0.1700

Data Index: 1628697 (Python Code)

Max Activating Token Index: 250

Text #7

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5455. Min Act: -0.1700

Data Index: 1621661 (Python Code)

Max Activating Token Index: 147

Text #8

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4828. Min Act: -0.1700

Data Index: 333649 (C4 (Web Text))

Max Activating Token Index: 1002

Text #9

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5141. Min Act: -0.1700

Data Index: 1615247 (Python Code)

Max Activating Token Index: 541

Text #10

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5141. Min Act: -0.1700

Data Index: 1419220 (Python Code)

Max Activating Token Index: 149

Text #11

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5141. Min Act: -0.1700

Data Index: 1310340 (C4 (Web Text))

Max Activating Token Index: 24

Text #12

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5455. Min Act: -0.1700

Data Index: 1267119 (C4 (Web Text))

Max Activating Token Index: 68

Text #13

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5141. Min Act: -0.1700

Data Index: 418778 (C4 (Web Text))

Max Activating Token Index: 23

Text #14

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.5455. Min Act: -0.1700

Data Index: 134339 (C4 (Web Text))

Max Activating Token Index: 614

Text #15

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4828. Min Act: -0.1700

Data Index: 332291 (C4 (Web Text))

Max Activating Token Index: 931

Text #16

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4828. Min Act: -0.1700

Data Index: 515112 (C4 (Web Text))

Max Activating Token Index: 765

Text #17

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4515. Min Act: -0.1700

Data Index: 1447605 (Python Code)

Max Activating Token Index: 571

Text #18

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4828. Min Act: -0.1700

Data Index: 1548102 (Python Code)

Max Activating Token Index: 193

Text #19

Max Range: 3.7020. Min Range: -3.7020

Max Act: 3.4515. Min Act: -0.1700

Data Index: 1410581 (Python Code)

Max Activating Token Index: 390