Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3588 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 6.3253. Min Range: -6.3253

Max Act: 6.3253. Min Act: -0.0000

Data Index: 1370355 (Python Code)

Max Activating Token Index: 356

Text #1

Max Range: 6.3253. Min Range: -6.3253

Max Act: 5.8507. Min Act: -0.0000

Data Index: 649254 (C4 (Web Text))

Max Activating Token Index: 573

Text #2

Max Range: 6.3253. Min Range: -6.3253

Max Act: 5.2518. Min Act: -0.0000

Data Index: 958783 (C4 (Web Text))

Max Activating Token Index: 299

Text #3

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.7254. Min Act: -0.0000

Data Index: 354801 (C4 (Web Text))

Max Activating Token Index: 459

Text #4

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.9820. Min Act: -0.0000

Data Index: 1509530 (Python Code)

Max Activating Token Index: 686

Text #5

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.9319. Min Act: -0.0000

Data Index: 1615514 (Python Code)

Max Activating Token Index: 911

Text #6

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.9309. Min Act: -0.0000

Data Index: 1558989 (Python Code)

Max Activating Token Index: 463

Text #7

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.6310. Min Act: -0.0000

Data Index: 1419363 (Python Code)

Max Activating Token Index: 423

Text #8

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.4386. Min Act: -0.0000

Data Index: 1561707 (Python Code)

Max Activating Token Index: 10

Text #9

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.5186. Min Act: -0.0000

Data Index: 1646917 (Python Code)

Max Activating Token Index: 983

Text #10

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.9168. Min Act: -0.0000

Data Index: 104547 (C4 (Web Text))

Max Activating Token Index: 922

Text #11

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.4311. Min Act: -0.0000

Data Index: 1367615 (Python Code)

Max Activating Token Index: 455

Text #12

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.4209. Min Act: -0.0000

Data Index: 1452829 (Python Code)

Max Activating Token Index: 950

Text #13

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.7236. Min Act: -0.0000

Data Index: 1505237 (Python Code)

Max Activating Token Index: 967

Text #14

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.1925. Min Act: -0.0000

Data Index: 1556272 (Python Code)

Max Activating Token Index: 320

Text #15

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.3071. Min Act: -0.0000

Data Index: 796406 (C4 (Web Text))

Max Activating Token Index: 684

Text #16

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.3173. Min Act: -0.0000

Data Index: 1089297 (C4 (Web Text))

Max Activating Token Index: 212

Text #17

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.4362. Min Act: -0.0000

Data Index: 1514101 (Python Code)

Max Activating Token Index: 190

Text #18

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.2592. Min Act: -0.0000

Data Index: 1030201 (C4 (Web Text))

Max Activating Token Index: 305

Text #19

Max Range: 6.3253. Min Range: -6.3253

Max Act: 4.1372. Min Act: -0.0000

Data Index: 221662 (C4 (Web Text))

Max Activating Token Index: 674