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

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

Neuron 4047 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3265. Min Range: -3.3265

Max Act: 3.3265. Min Act: -0.0000

Data Index: 890425 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.9346. Min Act: -0.0000

Data Index: 88162 (C4 (Web Text))

Max Activating Token Index: 748

Text #2

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.5294. Min Act: -0.0000

Data Index: 366460 (C4 (Web Text))

Max Activating Token Index: 1008

Text #3

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.4977. Min Act: -0.0000

Data Index: 513740 (C4 (Web Text))

Max Activating Token Index: 934

Text #4

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.4619. Min Act: -0.0000

Data Index: 353858 (C4 (Web Text))

Max Activating Token Index: 839

Text #5

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.4148. Min Act: -0.0000

Data Index: 1008473 (C4 (Web Text))

Max Activating Token Index: 164

Text #6

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.3926. Min Act: -0.0000

Data Index: 445173 (C4 (Web Text))

Max Activating Token Index: 673

Text #7

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.2490. Min Act: -0.0000

Data Index: 851372 (C4 (Web Text))

Max Activating Token Index: 520

Text #8

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.2566. Min Act: -0.0000

Data Index: 330263 (C4 (Web Text))

Max Activating Token Index: 582

Text #9

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.3131. Min Act: -0.0000

Data Index: 1090330 (C4 (Web Text))

Max Activating Token Index: 939

Text #10

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.9221. Min Act: -0.0000

Data Index: 1651240 (Python Code)

Max Activating Token Index: 852

Text #11

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.2393. Min Act: -0.0000

Data Index: 446958 (C4 (Web Text))

Max Activating Token Index: 614

Text #12

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.0766. Min Act: -0.0000

Data Index: 1360444 (Python Code)

Max Activating Token Index: 950

Text #13

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.9576. Min Act: -0.0000

Data Index: 932393 (C4 (Web Text))

Max Activating Token Index: 373

Text #14

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.8871. Min Act: -0.0000

Data Index: 1029644 (C4 (Web Text))

Max Activating Token Index: 842

Text #15

Max Range: 3.3265. Min Range: -3.3265

Max Act: 2.0243. Min Act: -0.0000

Data Index: 1366557 (Python Code)

Max Activating Token Index: 59

Text #16

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.9019. Min Act: -0.0000

Data Index: 1076628 (C4 (Web Text))

Max Activating Token Index: 522

Text #17

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.8896. Min Act: -0.0000

Data Index: 67755 (C4 (Web Text))

Max Activating Token Index: 249

Text #18

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.9334. Min Act: -0.0000

Data Index: 1531627 (Python Code)

Max Activating Token Index: 708

Text #19

Max Range: 3.3265. Min Range: -3.3265

Max Act: 1.7833. Min Act: -0.0000

Data Index: 259191 (C4 (Web Text))

Max Activating Token Index: 383