Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 256 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0339. Min Act: -0.0003

Data Index: 925218 (C4 (Web Text))

Max Activating Token Index: 326

Text #1

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0312. Min Act: -0.0003

Data Index: 645550 (C4 (Web Text))

Max Activating Token Index: 667

Text #2

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0338. Min Act: -0.0003

Data Index: 1364798 (Python Code)

Max Activating Token Index: 263

Text #3

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0310. Min Act: -0.0003

Data Index: 419531 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0299. Min Act: -0.0003

Data Index: 278670 (C4 (Web Text))

Max Activating Token Index: 630

Text #5

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0301. Min Act: -0.0003

Data Index: 1147006 (C4 (Web Text))

Max Activating Token Index: 466

Text #6

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0294. Min Act: -0.0003

Data Index: 178424 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0311. Min Act: -0.0003

Data Index: 337205 (C4 (Web Text))

Max Activating Token Index: 340

Text #8

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0298. Min Act: -0.0003

Data Index: 42066 (C4 (Web Text))

Max Activating Token Index: 702

Text #9

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0303. Min Act: -0.0003

Data Index: 638413 (C4 (Web Text))

Max Activating Token Index: 90

Text #10

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0293. Min Act: -0.0003

Data Index: 761403 (C4 (Web Text))

Max Activating Token Index: 556

Text #11

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0287. Min Act: -0.0003

Data Index: 1151359 (C4 (Web Text))

Max Activating Token Index: 779

Text #12

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0275. Min Act: -0.0003

Data Index: 1617668 (Python Code)

Max Activating Token Index: 869

Text #13

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0298. Min Act: -0.0003

Data Index: 221384 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0278. Min Act: -0.0003

Data Index: 458123 (C4 (Web Text))

Max Activating Token Index: 254

Text #15

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0312. Min Act: -0.0003

Data Index: 838495 (C4 (Web Text))

Max Activating Token Index: 911

Text #16

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0296. Min Act: -0.0003

Data Index: 918329 (C4 (Web Text))

Max Activating Token Index: 565

Text #17

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0288. Min Act: -0.0003

Data Index: 1319944 (C4 (Web Text))

Max Activating Token Index: 180

Text #18

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0288. Min Act: -0.0003

Data Index: 903011 (C4 (Web Text))

Max Activating Token Index: 266

Text #19

Max Range: 0.0339. Min Range: -0.0339

Max Act: 0.0281. Min Act: -0.0003

Data Index: 1319213 (C4 (Web Text))

Max Activating Token Index: 15