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

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

Neuron 4651 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.2437. Min Range: -4.2437

Max Act: 4.2437. Min Act: -0.0000

Data Index: 524979 (C4 (Web Text))

Max Activating Token Index: 399

Text #1

Max Range: 4.2437. Min Range: -4.2437

Max Act: 4.0367. Min Act: -0.0000

Data Index: 270540 (C4 (Web Text))

Max Activating Token Index: 38

Text #2

Max Range: 4.2437. Min Range: -4.2437

Max Act: 3.5693. Min Act: -0.0000

Data Index: 39512 (C4 (Web Text))

Max Activating Token Index: 335

Text #3

Max Range: 4.2437. Min Range: -4.2437

Max Act: 3.4689. Min Act: -0.0001

Data Index: 337945 (C4 (Web Text))

Max Activating Token Index: 841

Text #4

Max Range: 4.2437. Min Range: -4.2437

Max Act: 3.2151. Min Act: -0.0000

Data Index: 787367 (C4 (Web Text))

Max Activating Token Index: 4

Text #5

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.9840. Min Act: -0.0000

Data Index: 896174 (C4 (Web Text))

Max Activating Token Index: 70

Text #6

Max Range: 4.2437. Min Range: -4.2437

Max Act: 3.1859. Min Act: -0.0000

Data Index: 149687 (C4 (Web Text))

Max Activating Token Index: 120

Text #7

Max Range: 4.2437. Min Range: -4.2437

Max Act: 3.0072. Min Act: -0.0000

Data Index: 287727 (C4 (Web Text))

Max Activating Token Index: 475

Text #8

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.6054. Min Act: -0.0000

Data Index: 574858 (C4 (Web Text))

Max Activating Token Index: 998

Text #9

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.7050. Min Act: -0.0000

Data Index: 440137 (C4 (Web Text))

Max Activating Token Index: 604

Text #10

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.3157. Min Act: -0.0000

Data Index: 773923 (C4 (Web Text))

Max Activating Token Index: 166

Text #11

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.5621. Min Act: -0.0000

Data Index: 1294598 (C4 (Web Text))

Max Activating Token Index: 962

Text #12

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.4837. Min Act: -0.0000

Data Index: 381502 (C4 (Web Text))

Max Activating Token Index: 444

Text #13

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.6192. Min Act: -0.0000

Data Index: 985563 (C4 (Web Text))

Max Activating Token Index: 55

Text #14

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.4493. Min Act: -0.0000

Data Index: 882508 (C4 (Web Text))

Max Activating Token Index: 317

Text #15

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.3364. Min Act: -0.0000

Data Index: 1038167 (C4 (Web Text))

Max Activating Token Index: 860

Text #16

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.2542. Min Act: -0.0000

Data Index: 773709 (C4 (Web Text))

Max Activating Token Index: 322

Text #17

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.3646. Min Act: -0.0000

Data Index: 318756 (C4 (Web Text))

Max Activating Token Index: 267

Text #18

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.2326. Min Act: -0.0000

Data Index: 800001 (C4 (Web Text))

Max Activating Token Index: 578

Text #19

Max Range: 4.2437. Min Range: -4.2437

Max Act: 2.4714. Min Act: -0.0000

Data Index: 831103 (C4 (Web Text))

Max Activating Token Index: 96