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

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

Neuron 3642 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.4504. Min Range: -4.4504

Max Act: 4.4504. Min Act: -0.0000

Data Index: 938035 (C4 (Web Text))

Max Activating Token Index: 107

Text #1

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.6956. Min Act: -0.0000

Data Index: 266866 (C4 (Web Text))

Max Activating Token Index: 221

Text #2

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.8491. Min Act: -0.0000

Data Index: 10016 (C4 (Web Text))

Max Activating Token Index: 281

Text #3

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.4820. Min Act: -0.0000

Data Index: 1098282 (C4 (Web Text))

Max Activating Token Index: 842

Text #4

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.0644. Min Act: -0.0000

Data Index: 810707 (C4 (Web Text))

Max Activating Token Index: 963

Text #5

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.9909. Min Act: -0.0000

Data Index: 116565 (C4 (Web Text))

Max Activating Token Index: 16

Text #6

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.2689. Min Act: -0.0000

Data Index: 328160 (C4 (Web Text))

Max Activating Token Index: 471

Text #7

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.5950. Min Act: -0.0000

Data Index: 658373 (C4 (Web Text))

Max Activating Token Index: 684

Text #8

Max Range: 4.4504. Min Range: -4.4504

Max Act: 3.0861. Min Act: -0.0000

Data Index: 1162443 (C4 (Web Text))

Max Activating Token Index: 272

Text #9

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.8983. Min Act: -0.0000

Data Index: 1643674 (Python Code)

Max Activating Token Index: 286

Text #10

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.6520. Min Act: -0.0000

Data Index: 69559 (C4 (Web Text))

Max Activating Token Index: 106

Text #11

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.5234. Min Act: -0.0000

Data Index: 513599 (C4 (Web Text))

Max Activating Token Index: 211

Text #12

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.7633. Min Act: -0.0000

Data Index: 32926 (C4 (Web Text))

Max Activating Token Index: 43

Text #13

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.4650. Min Act: -0.0000

Data Index: 636717 (C4 (Web Text))

Max Activating Token Index: 205

Text #14

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.4119. Min Act: -0.0000

Data Index: 217173 (C4 (Web Text))

Max Activating Token Index: 898

Text #15

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.5659. Min Act: -0.0000

Data Index: 1283321 (C4 (Web Text))

Max Activating Token Index: 932

Text #16

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.5318. Min Act: -0.0000

Data Index: 1353631 (C4 (Web Text))

Max Activating Token Index: 783

Text #17

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.7734. Min Act: -0.0000

Data Index: 1389074 (Python Code)

Max Activating Token Index: 753

Text #18

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.6137. Min Act: -0.0000

Data Index: 1032769 (C4 (Web Text))

Max Activating Token Index: 424

Text #19

Max Range: 4.4504. Min Range: -4.4504

Max Act: 2.6207. Min Act: -0.0000

Data Index: 707833 (C4 (Web Text))

Max Activating Token Index: 11