Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2396 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 7.1839. Min Range: -7.1839

Max Act: 7.1839. Min Act: -0.0001

Data Index: 586679 (C4 (Web Text))

Max Activating Token Index: 132

Text #1

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.2895. Min Act: -0.0000

Data Index: 1318851 (C4 (Web Text))

Max Activating Token Index: 386

Text #2

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.2578. Min Act: -0.0000

Data Index: 1162003 (C4 (Web Text))

Max Activating Token Index: 544

Text #3

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.3616. Min Act: -0.0001

Data Index: 1137077 (C4 (Web Text))

Max Activating Token Index: 222

Text #4

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.2556. Min Act: -0.0001

Data Index: 766155 (C4 (Web Text))

Max Activating Token Index: 55

Text #5

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.9152. Min Act: -0.0001

Data Index: 377606 (C4 (Web Text))

Max Activating Token Index: 1021

Text #6

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.3893. Min Act: -0.0001

Data Index: 292780 (C4 (Web Text))

Max Activating Token Index: 108

Text #7

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.0299. Min Act: -0.0001

Data Index: 1067857 (C4 (Web Text))

Max Activating Token Index: 347

Text #8

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.0246. Min Act: -0.0001

Data Index: 70873 (C4 (Web Text))

Max Activating Token Index: 453

Text #9

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.9641. Min Act: -0.0001

Data Index: 448745 (C4 (Web Text))

Max Activating Token Index: 376

Text #10

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.7826. Min Act: -0.0001

Data Index: 233996 (C4 (Web Text))

Max Activating Token Index: 765

Text #11

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.7574. Min Act: -0.0001

Data Index: 939700 (C4 (Web Text))

Max Activating Token Index: 124

Text #12

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.7760. Min Act: -0.0001

Data Index: 1480153 (Python Code)

Max Activating Token Index: 825

Text #13

Max Range: 7.1839. Min Range: -7.1839

Max Act: 6.0032. Min Act: -0.0001

Data Index: 742803 (C4 (Web Text))

Max Activating Token Index: 980

Text #14

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.6446. Min Act: -0.0001

Data Index: 1430961 (Python Code)

Max Activating Token Index: 1010

Text #15

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.9638. Min Act: -0.0001

Data Index: 166579 (C4 (Web Text))

Max Activating Token Index: 204

Text #16

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.3948. Min Act: -0.0001

Data Index: 1177901 (C4 (Web Text))

Max Activating Token Index: 70

Text #17

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.5052. Min Act: -0.0001

Data Index: 498277 (C4 (Web Text))

Max Activating Token Index: 249

Text #18

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.4580. Min Act: -0.0000

Data Index: 1075807 (C4 (Web Text))

Max Activating Token Index: 493

Text #19

Max Range: 7.1839. Min Range: -7.1839

Max Act: 5.7388. Min Act: -0.0001

Data Index: 658580 (C4 (Web Text))

Max Activating Token Index: 585