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

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

Neuron 175 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: 3.2949. Min Range: -3.2949

Max Act: 3.2949. Min Act: -0.1700

Data Index: 986911 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.2320. Min Act: -0.1700

Data Index: 1382908 (Python Code)

Max Activating Token Index: 757

Text #2

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.2163. Min Act: -0.1700

Data Index: 34665 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1533. Min Act: -0.1700

Data Index: 497478 (C4 (Web Text))

Max Activating Token Index: 531

Text #4

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1691. Min Act: -0.1700

Data Index: 517249 (C4 (Web Text))

Max Activating Token Index: 875

Text #5

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1848. Min Act: -0.1700

Data Index: 1525842 (Python Code)

Max Activating Token Index: 780

Text #6

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1218. Min Act: -0.1700

Data Index: 588723 (C4 (Web Text))

Max Activating Token Index: 747

Text #7

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1061. Min Act: -0.1700

Data Index: 1211992 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1061. Min Act: -0.1700

Data Index: 131827 (C4 (Web Text))

Max Activating Token Index: 201

Text #9

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.1061. Min Act: -0.1700

Data Index: 58048 (C4 (Web Text))

Max Activating Token Index: 823

Text #10

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0903. Min Act: -0.1700

Data Index: 1321156 (C4 (Web Text))

Max Activating Token Index: 206

Text #11

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0429. Min Act: -0.1700

Data Index: 636200 (C4 (Web Text))

Max Activating Token Index: 139

Text #12

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0429. Min Act: -0.1700

Data Index: 1599555 (Python Code)

Max Activating Token Index: 259

Text #13

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0272. Min Act: -0.1700

Data Index: 599743 (C4 (Web Text))

Max Activating Token Index: 444

Text #14

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0114. Min Act: -0.1700

Data Index: 380468 (C4 (Web Text))

Max Activating Token Index: 680

Text #15

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0272. Min Act: -0.1700

Data Index: 1386728 (Python Code)

Max Activating Token Index: 331

Text #16

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0272. Min Act: -0.1700

Data Index: 1383294 (Python Code)

Max Activating Token Index: 807

Text #17

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0429. Min Act: -0.1700

Data Index: 900376 (C4 (Web Text))

Max Activating Token Index: 627

Text #18

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0429. Min Act: -0.1700

Data Index: 1215424 (C4 (Web Text))

Max Activating Token Index: 761

Text #19

Max Range: 3.2949. Min Range: -3.2949

Max Act: 3.0114. Min Act: -0.1700

Data Index: 55183 (C4 (Web Text))

Max Activating Token Index: 305