Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 808 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0553. Min Act: -0.0002

Data Index: 1000281 (C4 (Web Text))

Max Activating Token Index: 727

Text #1

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0600. Min Act: -0.0002

Data Index: 502507 (C4 (Web Text))

Max Activating Token Index: 929

Text #2

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0458. Min Act: -0.0002

Data Index: 652566 (C4 (Web Text))

Max Activating Token Index: 132

Text #3

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0505. Min Act: -0.0002

Data Index: 716887 (C4 (Web Text))

Max Activating Token Index: 58

Text #4

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0557. Min Act: -0.0002

Data Index: 592791 (C4 (Web Text))

Max Activating Token Index: 306

Text #5

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0397. Min Act: -0.0002

Data Index: 578157 (C4 (Web Text))

Max Activating Token Index: 149

Text #6

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0496. Min Act: -0.0002

Data Index: 583156 (C4 (Web Text))

Max Activating Token Index: 10

Text #7

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0509. Min Act: -0.0002

Data Index: 52244 (C4 (Web Text))

Max Activating Token Index: 292

Text #8

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0465. Min Act: -0.0002

Data Index: 1374226 (Python Code)

Max Activating Token Index: 239

Text #9

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0475. Min Act: -0.0002

Data Index: 149895 (C4 (Web Text))

Max Activating Token Index: 56

Text #10

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0477. Min Act: -0.0002

Data Index: 1497955 (Python Code)

Max Activating Token Index: 619

Text #11

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0469. Min Act: -0.0002

Data Index: 1222245 (C4 (Web Text))

Max Activating Token Index: 80

Text #12

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0415. Min Act: -0.0002

Data Index: 1146720 (C4 (Web Text))

Max Activating Token Index: 240

Text #13

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0457. Min Act: -0.0002

Data Index: 1607755 (Python Code)

Max Activating Token Index: 564

Text #14

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0494. Min Act: -0.0002

Data Index: 433362 (C4 (Web Text))

Max Activating Token Index: 263

Text #15

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0440. Min Act: -0.0002

Data Index: 826440 (C4 (Web Text))

Max Activating Token Index: 360

Text #16

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0452. Min Act: -0.0002

Data Index: 1322286 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0444. Min Act: -0.0002

Data Index: 982661 (C4 (Web Text))

Max Activating Token Index: 768

Text #18

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0361. Min Act: -0.0002

Data Index: 676431 (C4 (Web Text))

Max Activating Token Index: 253

Text #19

Max Range: 0.0600. Min Range: -0.0600

Max Act: 0.0468. Min Act: -0.0002

Data Index: 1185436 (C4 (Web Text))

Max Activating Token Index: 10