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

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

Neuron 2236 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1770. Min Act: -0.0001

Data Index: 51168 (C4 (Web Text))

Max Activating Token Index: 18

Text #1

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1667. Min Act: -0.0001

Data Index: 82092 (C4 (Web Text))

Max Activating Token Index: 880

Text #2

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1670. Min Act: -0.0001

Data Index: 756883 (C4 (Web Text))

Max Activating Token Index: 69

Text #3

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1454. Min Act: -0.0001

Data Index: 1124216 (C4 (Web Text))

Max Activating Token Index: 222

Text #4

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1434. Min Act: -0.0002

Data Index: 1117369 (C4 (Web Text))

Max Activating Token Index: 994

Text #5

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1376. Min Act: -0.0002

Data Index: 26102 (C4 (Web Text))

Max Activating Token Index: 903

Text #6

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1339. Min Act: -0.0001

Data Index: 1323310 (C4 (Web Text))

Max Activating Token Index: 751

Text #7

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1266. Min Act: -0.0002

Data Index: 951630 (C4 (Web Text))

Max Activating Token Index: 622

Text #8

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1380. Min Act: -0.0002

Data Index: 957386 (C4 (Web Text))

Max Activating Token Index: 98

Text #9

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1253. Min Act: -0.0002

Data Index: 808983 (C4 (Web Text))

Max Activating Token Index: 631

Text #10

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1363. Min Act: -0.0001

Data Index: 481009 (C4 (Web Text))

Max Activating Token Index: 908

Text #11

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1309. Min Act: -0.0002

Data Index: 394693 (C4 (Web Text))

Max Activating Token Index: 189

Text #12

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1260. Min Act: -0.0002

Data Index: 1029527 (C4 (Web Text))

Max Activating Token Index: 502

Text #13

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1342. Min Act: -0.0002

Data Index: 569653 (C4 (Web Text))

Max Activating Token Index: 969

Text #14

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1270. Min Act: -0.0001

Data Index: 1105976 (C4 (Web Text))

Max Activating Token Index: 884

Text #15

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1356. Min Act: -0.0001

Data Index: 733390 (C4 (Web Text))

Max Activating Token Index: 202

Text #16

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1210. Min Act: -0.0001

Data Index: 1375089 (Python Code)

Max Activating Token Index: 14

Text #17

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1242. Min Act: -0.0001

Data Index: 891395 (C4 (Web Text))

Max Activating Token Index: 478

Text #18

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1263. Min Act: -0.0001

Data Index: 866597 (C4 (Web Text))

Max Activating Token Index: 803

Text #19

Max Range: 0.1770. Min Range: -0.1770

Max Act: 0.1212. Min Act: -0.0002

Data Index: 1225200 (C4 (Web Text))

Max Activating Token Index: 989