Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1536 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1778. Min Act: -0.0003

Data Index: 6842 (C4 (Web Text))

Max Activating Token Index: 63

Text #1

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1564. Min Act: -0.0003

Data Index: 516006 (C4 (Web Text))

Max Activating Token Index: 34

Text #2

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1585. Min Act: -0.0003

Data Index: 1083724 (C4 (Web Text))

Max Activating Token Index: 170

Text #3

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1307. Min Act: -0.0003

Data Index: 81999 (C4 (Web Text))

Max Activating Token Index: 664

Text #4

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1346. Min Act: -0.0003

Data Index: 388938 (C4 (Web Text))

Max Activating Token Index: 851

Text #5

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1320. Min Act: -0.0004

Data Index: 360337 (C4 (Web Text))

Max Activating Token Index: 585

Text #6

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1440. Min Act: -0.0003

Data Index: 356645 (C4 (Web Text))

Max Activating Token Index: 174

Text #7

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1320. Min Act: -0.0003

Data Index: 397240 (C4 (Web Text))

Max Activating Token Index: 522

Text #8

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1334. Min Act: -0.0003

Data Index: 81198 (C4 (Web Text))

Max Activating Token Index: 324

Text #9

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1301. Min Act: -0.0003

Data Index: 1106152 (C4 (Web Text))

Max Activating Token Index: 332

Text #10

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1288. Min Act: -0.0004

Data Index: 841097 (C4 (Web Text))

Max Activating Token Index: 900

Text #11

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1345. Min Act: -0.0003

Data Index: 976943 (C4 (Web Text))

Max Activating Token Index: 361

Text #12

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1324. Min Act: -0.0004

Data Index: 892792 (C4 (Web Text))

Max Activating Token Index: 518

Text #13

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1311. Min Act: -0.0003

Data Index: 378166 (C4 (Web Text))

Max Activating Token Index: 398

Text #14

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1316. Min Act: -0.0003

Data Index: 874755 (C4 (Web Text))

Max Activating Token Index: 144

Text #15

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1321. Min Act: -0.0003

Data Index: 767116 (C4 (Web Text))

Max Activating Token Index: 127

Text #16

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1354. Min Act: -0.0003

Data Index: 34507 (C4 (Web Text))

Max Activating Token Index: 571

Text #17

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1311. Min Act: -0.0003

Data Index: 203847 (C4 (Web Text))

Max Activating Token Index: 67

Text #18

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1338. Min Act: -0.0003

Data Index: 941458 (C4 (Web Text))

Max Activating Token Index: 651

Text #19

Max Range: 0.1778. Min Range: -0.1778

Max Act: 0.1302. Min Act: -0.0003

Data Index: 774663 (C4 (Web Text))

Max Activating Token Index: 254