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

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

Neuron 4976 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.3037. Min Act: -0.0000

Data Index: 34001 (C4 (Web Text))

Max Activating Token Index: 170

Text #1

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2882. Min Act: -0.0000

Data Index: 1284024 (C4 (Web Text))

Max Activating Token Index: 642

Text #2

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2634. Min Act: -0.0000

Data Index: 395166 (C4 (Web Text))

Max Activating Token Index: 960

Text #3

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2572. Min Act: -0.0000

Data Index: 1047352 (C4 (Web Text))

Max Activating Token Index: 266

Text #4

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2532. Min Act: -0.0000

Data Index: 313246 (C4 (Web Text))

Max Activating Token Index: 798

Text #5

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2531. Min Act: -0.0000

Data Index: 764421 (C4 (Web Text))

Max Activating Token Index: 610

Text #6

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2446. Min Act: -0.0000

Data Index: 609879 (C4 (Web Text))

Max Activating Token Index: 289

Text #7

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2475. Min Act: -0.0000

Data Index: 806329 (C4 (Web Text))

Max Activating Token Index: 816

Text #8

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2313. Min Act: -0.0000

Data Index: 1334077 (C4 (Web Text))

Max Activating Token Index: 542

Text #9

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2347. Min Act: -0.0000

Data Index: 253796 (C4 (Web Text))

Max Activating Token Index: 188

Text #10

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2460. Min Act: -0.0000

Data Index: 377721 (C4 (Web Text))

Max Activating Token Index: 417

Text #11

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2375. Min Act: -0.0000

Data Index: 1075036 (C4 (Web Text))

Max Activating Token Index: 691

Text #12

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2441. Min Act: -0.0000

Data Index: 383460 (C4 (Web Text))

Max Activating Token Index: 532

Text #13

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2231. Min Act: -0.0000

Data Index: 1044431 (C4 (Web Text))

Max Activating Token Index: 810

Text #14

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2066. Min Act: -0.0000

Data Index: 931763 (C4 (Web Text))

Max Activating Token Index: 304

Text #15

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2284. Min Act: -0.0000

Data Index: 800345 (C4 (Web Text))

Max Activating Token Index: 340

Text #16

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2255. Min Act: -0.0000

Data Index: 1582612 (Python Code)

Max Activating Token Index: 366

Text #17

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2243. Min Act: -0.0000

Data Index: 1487912 (Python Code)

Max Activating Token Index: 429

Text #18

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2156. Min Act: -0.0000

Data Index: 1560248 (Python Code)

Max Activating Token Index: 311

Text #19

Max Range: 0.3037. Min Range: -0.3037

Max Act: 0.2235. Min Act: -0.0000

Data Index: 1499680 (Python Code)

Max Activating Token Index: 426