Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1795 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.3860. Min Act: -0.0001

Data Index: 1180112 (C4 (Web Text))

Max Activating Token Index: 832

Text #1

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.3587. Min Act: -0.0001

Data Index: 257933 (C4 (Web Text))

Max Activating Token Index: 629

Text #2

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.2949. Min Act: -0.0001

Data Index: 941706 (C4 (Web Text))

Max Activating Token Index: 684

Text #3

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.3452. Min Act: -0.0001

Data Index: 1622704 (Python Code)

Max Activating Token Index: 265

Text #4

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1586. Min Act: -0.0001

Data Index: 554974 (C4 (Web Text))

Max Activating Token Index: 188

Text #5

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1796. Min Act: -0.0001

Data Index: 93108 (C4 (Web Text))

Max Activating Token Index: 332

Text #6

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1180. Min Act: -0.0001

Data Index: 682589 (C4 (Web Text))

Max Activating Token Index: 181

Text #7

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1778. Min Act: -0.0001

Data Index: 1492192 (Python Code)

Max Activating Token Index: 195

Text #8

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.2898. Min Act: -0.0001

Data Index: 637820 (C4 (Web Text))

Max Activating Token Index: 874

Text #9

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1173. Min Act: -0.0001

Data Index: 565264 (C4 (Web Text))

Max Activating Token Index: 119

Text #10

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0434. Min Act: -0.0001

Data Index: 384679 (C4 (Web Text))

Max Activating Token Index: 950

Text #11

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.1706. Min Act: -0.0001

Data Index: 1269335 (C4 (Web Text))

Max Activating Token Index: 742

Text #12

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0986. Min Act: -0.0001

Data Index: 784990 (C4 (Web Text))

Max Activating Token Index: 451

Text #13

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0974. Min Act: -0.0001

Data Index: 1435369 (Python Code)

Max Activating Token Index: 29

Text #14

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0984. Min Act: -0.0001

Data Index: 472576 (C4 (Web Text))

Max Activating Token Index: 987

Text #15

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0114. Min Act: -0.0001

Data Index: 1585192 (Python Code)

Max Activating Token Index: 337

Text #16

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0115. Min Act: -0.0001

Data Index: 1587266 (Python Code)

Max Activating Token Index: 807

Text #17

Max Range: 1.3860. Min Range: -1.3860

Max Act: 0.9995. Min Act: -0.0001

Data Index: 581528 (C4 (Web Text))

Max Activating Token Index: 536

Text #18

Max Range: 1.3860. Min Range: -1.3860

Max Act: 0.9815. Min Act: -0.0001

Data Index: 325147 (C4 (Web Text))

Max Activating Token Index: 474

Text #19

Max Range: 1.3860. Min Range: -1.3860

Max Act: 1.0356. Min Act: -0.0001

Data Index: 644536 (C4 (Web Text))

Max Activating Token Index: 680