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

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

Neuron 1561 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: 5.2092. Min Range: -5.2092

Max Act: 5.1779. Min Act: -0.1700

Data Index: 725402 (C4 (Web Text))

Max Activating Token Index: 428

Text #1

Max Range: 5.2092. Min Range: -5.2092

Max Act: 5.2092. Min Act: -0.1700

Data Index: 856989 (C4 (Web Text))

Max Activating Token Index: 1009

Text #2

Max Range: 5.2092. Min Range: -5.2092

Max Act: 5.1779. Min Act: -0.1700

Data Index: 471042 (C4 (Web Text))

Max Activating Token Index: 1021

Text #3

Max Range: 5.2092. Min Range: -5.2092

Max Act: 5.0529. Min Act: -0.1700

Data Index: 1172969 (C4 (Web Text))

Max Activating Token Index: 73

Text #4

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.9279. Min Act: -0.1700

Data Index: 1308729 (C4 (Web Text))

Max Activating Token Index: 868

Text #5

Max Range: 5.2092. Min Range: -5.2092

Max Act: 5.0217. Min Act: -0.1700

Data Index: 417647 (C4 (Web Text))

Max Activating Token Index: 827

Text #6

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.9279. Min Act: -0.1700

Data Index: 516432 (C4 (Web Text))

Max Activating Token Index: 658

Text #7

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.9592. Min Act: -0.1700

Data Index: 888493 (C4 (Web Text))

Max Activating Token Index: 813

Text #8

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.8967. Min Act: -0.1700

Data Index: 194985 (C4 (Web Text))

Max Activating Token Index: 957

Text #9

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.8029. Min Act: -0.1700

Data Index: 326544 (C4 (Web Text))

Max Activating Token Index: 513

Text #10

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7404. Min Act: -0.1700

Data Index: 436844 (C4 (Web Text))

Max Activating Token Index: 550

Text #11

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.8342. Min Act: -0.1700

Data Index: 900941 (C4 (Web Text))

Max Activating Token Index: 535

Text #12

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7717. Min Act: -0.1700

Data Index: 1205872 (C4 (Web Text))

Max Activating Token Index: 790

Text #13

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7404. Min Act: -0.1700

Data Index: 1045313 (C4 (Web Text))

Max Activating Token Index: 245

Text #14

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7404. Min Act: -0.1700

Data Index: 565052 (C4 (Web Text))

Max Activating Token Index: 924

Text #15

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7092. Min Act: -0.1700

Data Index: 993407 (C4 (Web Text))

Max Activating Token Index: 232

Text #16

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7404. Min Act: -0.1700

Data Index: 1436043 (Python Code)

Max Activating Token Index: 994

Text #17

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.6779. Min Act: -0.1700

Data Index: 264871 (C4 (Web Text))

Max Activating Token Index: 891

Text #18

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.7404. Min Act: -0.1700

Data Index: 444559 (C4 (Web Text))

Max Activating Token Index: 683

Text #19

Max Range: 5.2092. Min Range: -5.2092

Max Act: 4.6467. Min Act: -0.1700

Data Index: 800915 (C4 (Web Text))

Max Activating Token Index: 136