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

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

Neuron 1767 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: 5.2355. Min Range: -5.2355

Max Act: 5.2355. Min Act: -0.1700

Data Index: 551355 (C4 (Web Text))

Max Activating Token Index: 635

Text #1

Max Range: 5.2355. Min Range: -5.2355

Max Act: 5.0793. Min Act: -0.1700

Data Index: 727002 (C4 (Web Text))

Max Activating Token Index: 981

Text #2

Max Range: 5.2355. Min Range: -5.2355

Max Act: 5.0168. Min Act: -0.1700

Data Index: 680915 (C4 (Web Text))

Max Activating Token Index: 667

Text #3

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9855. Min Act: -0.1700

Data Index: 701271 (C4 (Web Text))

Max Activating Token Index: 340

Text #4

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9543. Min Act: -0.1700

Data Index: 1319978 (C4 (Web Text))

Max Activating Token Index: 426

Text #5

Max Range: 5.2355. Min Range: -5.2355

Max Act: 5.0168. Min Act: -0.1700

Data Index: 1308672 (C4 (Web Text))

Max Activating Token Index: 147

Text #6

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9543. Min Act: -0.1700

Data Index: 801893 (C4 (Web Text))

Max Activating Token Index: 896

Text #7

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9543. Min Act: -0.1700

Data Index: 108173 (C4 (Web Text))

Max Activating Token Index: 200

Text #8

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8918. Min Act: -0.1700

Data Index: 645382 (C4 (Web Text))

Max Activating Token Index: 575

Text #9

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9855. Min Act: -0.1700

Data Index: 401411 (C4 (Web Text))

Max Activating Token Index: 914

Text #10

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9543. Min Act: -0.1700

Data Index: 264069 (C4 (Web Text))

Max Activating Token Index: 116

Text #11

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8918. Min Act: -0.1700

Data Index: 235799 (C4 (Web Text))

Max Activating Token Index: 185

Text #12

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9230. Min Act: -0.1700

Data Index: 770471 (C4 (Web Text))

Max Activating Token Index: 839

Text #13

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9230. Min Act: -0.1700

Data Index: 308250 (C4 (Web Text))

Max Activating Token Index: 994

Text #14

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.9230. Min Act: -0.1700

Data Index: 853262 (C4 (Web Text))

Max Activating Token Index: 100

Text #15

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8918. Min Act: -0.1700

Data Index: 697977 (C4 (Web Text))

Max Activating Token Index: 693

Text #16

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8293. Min Act: -0.1700

Data Index: 948332 (C4 (Web Text))

Max Activating Token Index: 358

Text #17

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8918. Min Act: -0.1700

Data Index: 60965 (C4 (Web Text))

Max Activating Token Index: 701

Text #18

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8605. Min Act: -0.1700

Data Index: 519580 (C4 (Web Text))

Max Activating Token Index: 107

Text #19

Max Range: 5.2355. Min Range: -5.2355

Max Act: 4.8605. Min Act: -0.1700

Data Index: 128111 (C4 (Web Text))

Max Activating Token Index: 703