Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3874 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.4572. Min Act: -0.0000

Data Index: 56461 (C4 (Web Text))

Max Activating Token Index: 118

Text #1

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.3828. Min Act: -0.0000

Data Index: 204388 (C4 (Web Text))

Max Activating Token Index: 543

Text #2

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.3341. Min Act: -0.0000

Data Index: 1296578 (C4 (Web Text))

Max Activating Token Index: 419

Text #3

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.2607. Min Act: -0.0000

Data Index: 976586 (C4 (Web Text))

Max Activating Token Index: 345

Text #4

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.2493. Min Act: -0.0000

Data Index: 942198 (C4 (Web Text))

Max Activating Token Index: 1020

Text #5

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.3706. Min Act: -0.0000

Data Index: 1380354 (Python Code)

Max Activating Token Index: 174

Text #6

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.2207. Min Act: -0.0000

Data Index: 232566 (C4 (Web Text))

Max Activating Token Index: 54

Text #7

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1487. Min Act: -0.0000

Data Index: 588827 (C4 (Web Text))

Max Activating Token Index: 1010

Text #8

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.2660. Min Act: -0.0000

Data Index: 787376 (C4 (Web Text))

Max Activating Token Index: 748

Text #9

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1922. Min Act: -0.0000

Data Index: 213026 (C4 (Web Text))

Max Activating Token Index: 412

Text #10

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.2558. Min Act: -0.0000

Data Index: 1480241 (Python Code)

Max Activating Token Index: 693

Text #11

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1958. Min Act: -0.0000

Data Index: 821027 (C4 (Web Text))

Max Activating Token Index: 683

Text #12

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.0783. Min Act: -0.0000

Data Index: 1476979 (Python Code)

Max Activating Token Index: 1012

Text #13

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1820. Min Act: -0.0000

Data Index: 19695 (C4 (Web Text))

Max Activating Token Index: 321

Text #14

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1066. Min Act: -0.0000

Data Index: 1149638 (C4 (Web Text))

Max Activating Token Index: 630

Text #15

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1039. Min Act: -0.0000

Data Index: 314348 (C4 (Web Text))

Max Activating Token Index: 297

Text #16

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1623. Min Act: -0.0000

Data Index: 537187 (C4 (Web Text))

Max Activating Token Index: 306

Text #17

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1337. Min Act: -0.0000

Data Index: 991123 (C4 (Web Text))

Max Activating Token Index: 942

Text #18

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1684. Min Act: -0.0000

Data Index: 433038 (C4 (Web Text))

Max Activating Token Index: 865

Text #19

Max Range: 1.4572. Min Range: -1.4572

Max Act: 1.1144. Min Act: -0.0000

Data Index: 1095842 (C4 (Web Text))

Max Activating Token Index: 234