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

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

Neuron 1854 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.2670. Min Act: -0.0000

Data Index: 855284 (C4 (Web Text))

Max Activating Token Index: 247

Text #1

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.2555. Min Act: -0.0000

Data Index: 622212 (C4 (Web Text))

Max Activating Token Index: 453

Text #2

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.2451. Min Act: -0.0000

Data Index: 412082 (C4 (Web Text))

Max Activating Token Index: 523

Text #3

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.2325. Min Act: -0.0000

Data Index: 76814 (C4 (Web Text))

Max Activating Token Index: 504

Text #4

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.2231. Min Act: -0.0000

Data Index: 6680 (C4 (Web Text))

Max Activating Token Index: 237

Text #5

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1898. Min Act: -0.0000

Data Index: 979705 (C4 (Web Text))

Max Activating Token Index: 955

Text #6

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1904. Min Act: -0.0000

Data Index: 871439 (C4 (Web Text))

Max Activating Token Index: 241

Text #7

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1960. Min Act: -0.0000

Data Index: 153750 (C4 (Web Text))

Max Activating Token Index: 325

Text #8

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1856. Min Act: -0.0000

Data Index: 437093 (C4 (Web Text))

Max Activating Token Index: 93

Text #9

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1924. Min Act: -0.0000

Data Index: 811124 (C4 (Web Text))

Max Activating Token Index: 40

Text #10

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1933. Min Act: -0.0000

Data Index: 1005569 (C4 (Web Text))

Max Activating Token Index: 332

Text #11

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1796. Min Act: -0.0000

Data Index: 1499918 (Python Code)

Max Activating Token Index: 372

Text #12

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1724. Min Act: -0.0000

Data Index: 1410209 (Python Code)

Max Activating Token Index: 296

Text #13

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1784. Min Act: -0.0000

Data Index: 99108 (C4 (Web Text))

Max Activating Token Index: 685

Text #14

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1709. Min Act: -0.0000

Data Index: 596392 (C4 (Web Text))

Max Activating Token Index: 62

Text #15

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1768. Min Act: -0.0000

Data Index: 632170 (C4 (Web Text))

Max Activating Token Index: 166

Text #16

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1669. Min Act: -0.0000

Data Index: 553562 (C4 (Web Text))

Max Activating Token Index: 313

Text #17

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1720. Min Act: -0.0000

Data Index: 1477020 (Python Code)

Max Activating Token Index: 312

Text #18

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1588. Min Act: -0.0000

Data Index: 1466575 (Python Code)

Max Activating Token Index: 334

Text #19

Max Range: 0.2670. Min Range: -0.2670

Max Act: 0.1571. Min Act: -0.0000

Data Index: 678441 (C4 (Web Text))

Max Activating Token Index: 284