Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3496 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1803. Min Act: -0.0000

Data Index: 1466914 (Python Code)

Max Activating Token Index: 7

Text #1

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1649. Min Act: -0.0000

Data Index: 241270 (C4 (Web Text))

Max Activating Token Index: 819

Text #2

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1875. Min Act: -0.0000

Data Index: 150016 (C4 (Web Text))

Max Activating Token Index: 206

Text #3

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.0880. Min Act: -0.0000

Data Index: 1586217 (Python Code)

Max Activating Token Index: 10

Text #4

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1263. Min Act: -0.0000

Data Index: 97727 (C4 (Web Text))

Max Activating Token Index: 419

Text #5

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1142. Min Act: -0.0000

Data Index: 330102 (C4 (Web Text))

Max Activating Token Index: 680

Text #6

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.0112. Min Act: -0.0000

Data Index: 477323 (C4 (Web Text))

Max Activating Token Index: 466

Text #7

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.1034. Min Act: -0.0000

Data Index: 1016817 (C4 (Web Text))

Max Activating Token Index: 235

Text #8

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9348. Min Act: -0.0000

Data Index: 339544 (C4 (Web Text))

Max Activating Token Index: 265

Text #9

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.0807. Min Act: -0.0000

Data Index: 756103 (C4 (Web Text))

Max Activating Token Index: 909

Text #10

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9930. Min Act: -0.0000

Data Index: 570442 (C4 (Web Text))

Max Activating Token Index: 409

Text #11

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9312. Min Act: -0.0000

Data Index: 1415222 (Python Code)

Max Activating Token Index: 846

Text #12

Max Range: 1.1875. Min Range: -1.1875

Max Act: 1.0303. Min Act: -0.0000

Data Index: 82067 (C4 (Web Text))

Max Activating Token Index: 446

Text #13

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9469. Min Act: -0.0000

Data Index: 601757 (C4 (Web Text))

Max Activating Token Index: 575

Text #14

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9117. Min Act: -0.0000

Data Index: 637371 (C4 (Web Text))

Max Activating Token Index: 11

Text #15

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.8467. Min Act: -0.0000

Data Index: 1387599 (Python Code)

Max Activating Token Index: 448

Text #16

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9211. Min Act: -0.0000

Data Index: 615538 (C4 (Web Text))

Max Activating Token Index: 525

Text #17

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9658. Min Act: -0.0000

Data Index: 1332664 (C4 (Web Text))

Max Activating Token Index: 158

Text #18

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9282. Min Act: -0.0000

Data Index: 603089 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 1.1875. Min Range: -1.1875

Max Act: 0.9321. Min Act: -0.0000

Data Index: 1135891 (C4 (Web Text))

Max Activating Token Index: 672