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

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

Neuron 22 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.3069. Min Act: -0.0000

Data Index: 678442 (C4 (Web Text))

Max Activating Token Index: 294

Text #1

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2710. Min Act: -0.0000

Data Index: 1198508 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2545. Min Act: -0.0000

Data Index: 670525 (C4 (Web Text))

Max Activating Token Index: 999

Text #3

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2529. Min Act: -0.0000

Data Index: 57374 (C4 (Web Text))

Max Activating Token Index: 810

Text #4

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2508. Min Act: -0.0000

Data Index: 764227 (C4 (Web Text))

Max Activating Token Index: 185

Text #5

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2379. Min Act: -0.0000

Data Index: 1055557 (C4 (Web Text))

Max Activating Token Index: 384

Text #6

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2446. Min Act: -0.0000

Data Index: 1270008 (C4 (Web Text))

Max Activating Token Index: 515

Text #7

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2251. Min Act: -0.0000

Data Index: 514900 (C4 (Web Text))

Max Activating Token Index: 813

Text #8

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2475. Min Act: -0.0000

Data Index: 538641 (C4 (Web Text))

Max Activating Token Index: 632

Text #9

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2389. Min Act: -0.0000

Data Index: 250476 (C4 (Web Text))

Max Activating Token Index: 284

Text #10

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2414. Min Act: -0.0000

Data Index: 974260 (C4 (Web Text))

Max Activating Token Index: 797

Text #11

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2330. Min Act: -0.0000

Data Index: 507938 (C4 (Web Text))

Max Activating Token Index: 538

Text #12

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2471. Min Act: -0.0000

Data Index: 1071056 (C4 (Web Text))

Max Activating Token Index: 376

Text #13

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2299. Min Act: -0.0000

Data Index: 343476 (C4 (Web Text))

Max Activating Token Index: 522

Text #14

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2192. Min Act: -0.0000

Data Index: 1123407 (C4 (Web Text))

Max Activating Token Index: 727

Text #15

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2320. Min Act: -0.0000

Data Index: 1124428 (C4 (Web Text))

Max Activating Token Index: 323

Text #16

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2218. Min Act: -0.0000

Data Index: 1326007 (C4 (Web Text))

Max Activating Token Index: 356

Text #17

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2151. Min Act: -0.0000

Data Index: 830604 (C4 (Web Text))

Max Activating Token Index: 369

Text #18

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2155. Min Act: -0.0000

Data Index: 36499 (C4 (Web Text))

Max Activating Token Index: 1010

Text #19

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2097. Min Act: -0.0000

Data Index: 698671 (C4 (Web Text))

Max Activating Token Index: 626