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

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

Neuron 321 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: 0.0304. Min Range: -0.0304

Max Act: 0.0304. Min Act: -0.0002

Data Index: 99259 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0288. Min Act: -0.0003

Data Index: 267102 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0281. Min Act: -0.0002

Data Index: 1080313 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0290. Min Act: -0.0002

Data Index: 1226361 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0292. Min Act: -0.0003

Data Index: 1413393 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0284. Min Act: -0.0003

Data Index: 510161 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0282. Min Act: -0.0002

Data Index: 502377 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0279. Min Act: -0.0002

Data Index: 1384005 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0268. Min Act: -0.0002

Data Index: 425573 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0265. Min Act: -0.0003

Data Index: 11624 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0277. Min Act: -0.0003

Data Index: 1312114 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0279. Min Act: -0.0003

Data Index: 114192 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0270. Min Act: -0.0003

Data Index: 1218853 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0262. Min Act: -0.0003

Data Index: 1044396 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0263. Min Act: -0.0003

Data Index: 120216 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0271. Min Act: -0.0002

Data Index: 941535 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0274. Min Act: -0.0002

Data Index: 678525 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0266. Min Act: -0.0002

Data Index: 351821 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0267. Min Act: -0.0002

Data Index: 1571427 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 0.0304. Min Range: -0.0304

Max Act: 0.0256. Min Act: -0.0003

Data Index: 1057265 (C4 (Web Text))

Max Activating Token Index: 1023