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

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

Neuron 3811 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: 1.3165. Min Range: -1.3165

Max Act: 1.3165. Min Act: -0.0000

Data Index: 760133 (C4 (Web Text))

Max Activating Token Index: 465

Text #1

Max Range: 1.3165. Min Range: -1.3165

Max Act: 1.0098. Min Act: -0.0000

Data Index: 693605 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 1.3165. Min Range: -1.3165

Max Act: 1.0829. Min Act: -0.0000

Data Index: 642957 (C4 (Web Text))

Max Activating Token Index: 773

Text #3

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.9135. Min Act: -0.0000

Data Index: 602074 (C4 (Web Text))

Max Activating Token Index: 676

Text #4

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.9431. Min Act: -0.0000

Data Index: 618956 (C4 (Web Text))

Max Activating Token Index: 230

Text #5

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.9344. Min Act: -0.0000

Data Index: 339227 (C4 (Web Text))

Max Activating Token Index: 624

Text #6

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7904. Min Act: -0.0000

Data Index: 1062644 (C4 (Web Text))

Max Activating Token Index: 340

Text #7

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.9057. Min Act: -0.0000

Data Index: 283704 (C4 (Web Text))

Max Activating Token Index: 406

Text #8

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.8161. Min Act: -0.0000

Data Index: 720955 (C4 (Web Text))

Max Activating Token Index: 897

Text #9

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7482. Min Act: -0.0000

Data Index: 600801 (C4 (Web Text))

Max Activating Token Index: 89

Text #10

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.8168. Min Act: -0.0000

Data Index: 193948 (C4 (Web Text))

Max Activating Token Index: 789

Text #11

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7951. Min Act: -0.0000

Data Index: 442880 (C4 (Web Text))

Max Activating Token Index: 619

Text #12

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.8140. Min Act: -0.0000

Data Index: 926280 (C4 (Web Text))

Max Activating Token Index: 517

Text #13

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7547. Min Act: -0.0000

Data Index: 1369422 (Python Code)

Max Activating Token Index: 914

Text #14

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.8170. Min Act: -0.0000

Data Index: 320837 (C4 (Web Text))

Max Activating Token Index: 251

Text #15

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7381. Min Act: -0.0000

Data Index: 632723 (C4 (Web Text))

Max Activating Token Index: 415

Text #16

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7304. Min Act: -0.0000

Data Index: 1311912 (C4 (Web Text))

Max Activating Token Index: 273

Text #17

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7339. Min Act: -0.0000

Data Index: 1407690 (Python Code)

Max Activating Token Index: 481

Text #18

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7530. Min Act: -0.0000

Data Index: 1268463 (C4 (Web Text))

Max Activating Token Index: 149

Text #19

Max Range: 1.3165. Min Range: -1.3165

Max Act: 0.7056. Min Act: -0.0000

Data Index: 1531176 (Python Code)

Max Activating Token Index: 596