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

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

Neuron 770 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: 5.3820. Min Range: -5.3820

Max Act: 5.3820. Min Act: -0.1700

Data Index: 385863 (C4 (Web Text))

Max Activating Token Index: 557

Text #1

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.3507. Min Act: -0.1699

Data Index: 332743 (C4 (Web Text))

Max Activating Token Index: 527

Text #2

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.3195. Min Act: -0.1700

Data Index: 523862 (C4 (Web Text))

Max Activating Token Index: 1013

Text #3

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.3507. Min Act: -0.1700

Data Index: 168847 (C4 (Web Text))

Max Activating Token Index: 968

Text #4

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.2570. Min Act: -0.1700

Data Index: 50775 (C4 (Web Text))

Max Activating Token Index: 390

Text #5

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.2570. Min Act: -0.1700

Data Index: 546548 (C4 (Web Text))

Max Activating Token Index: 989

Text #6

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.2882. Min Act: -0.1700

Data Index: 150347 (C4 (Web Text))

Max Activating Token Index: 101

Text #7

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1945. Min Act: -0.1700

Data Index: 523490 (C4 (Web Text))

Max Activating Token Index: 583

Text #8

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1945. Min Act: -0.1700

Data Index: 110296 (C4 (Web Text))

Max Activating Token Index: 662

Text #9

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1945. Min Act: -0.1700

Data Index: 1302968 (C4 (Web Text))

Max Activating Token Index: 802

Text #10

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1632. Min Act: -0.1700

Data Index: 240900 (C4 (Web Text))

Max Activating Token Index: 866

Text #11

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1007. Min Act: -0.1700

Data Index: 315126 (C4 (Web Text))

Max Activating Token Index: 829

Text #12

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1320. Min Act: -0.1700

Data Index: 15858 (C4 (Web Text))

Max Activating Token Index: 589

Text #13

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1632. Min Act: -0.1700

Data Index: 1120777 (C4 (Web Text))

Max Activating Token Index: 911

Text #14

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1007. Min Act: -0.1700

Data Index: 1356084 (C4 (Web Text))

Max Activating Token Index: 421

Text #15

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.0382. Min Act: -0.1700

Data Index: 797911 (C4 (Web Text))

Max Activating Token Index: 221

Text #16

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.1007. Min Act: -0.1700

Data Index: 564582 (C4 (Web Text))

Max Activating Token Index: 254

Text #17

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.0695. Min Act: -0.1700

Data Index: 1193975 (C4 (Web Text))

Max Activating Token Index: 813

Text #18

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.0382. Min Act: -0.1700

Data Index: 1089682 (C4 (Web Text))

Max Activating Token Index: 600

Text #19

Max Range: 5.3820. Min Range: -5.3820

Max Act: 5.0070. Min Act: -0.1700

Data Index: 1049422 (C4 (Web Text))

Max Activating Token Index: 259