What does a random seed reproduce?

Reference inputs · synthetic data

Records
4
Data columns
2
Missing cells
0
Duplicate records
0
Feature duplicates
0

4 of 4 records · One record = one generator transition input. Statistics and downloads use all records.

About this dataset

Toy recurrence x ← (5x + 3) mod 16; u = x / 16. Each recorded run restarts at the chosen seed, consumes the selected number of earlier outputs, then shows the next outputs. Fixed category probabilities are 60%, 30%, 10%. This tiny 16-state generator illustrates state and repetition; it is not PyTorch’s generator and cannot represent those probabilities exactly.

  • This is a deterministic transition trace, not a sampled training dataset. The input to each recurrence is its previous state. All discarded transitions remain visible.
  • The toy recurrence has period 16 for every allowed seed; seeds choose starting positions on the same cycle. Uniform-grid values are not ideal continuous uniforms.
  • Fixed intervals [0,.6), [.6,.9), [.9,1) illustrate sampling. Of the 16 possible grid values, 10 map to a, 5 to b, and 1 to c; the toy therefore does not reproduce 60%,30%,10% exactly.
  • Reference inputs use seed 7 with no earlier consumption. Reports include actual run settings and every transition.

Counts exclude the row identifier. Duplicate counts are extra occurrences after the first. Feature duplicates ignore labels and experimental conditions. — means a statistic is undefined or not applicable.

Row
Stable record identifier, not a model feature.
State before transition
State before transition
Use of output
Use of output