← A tiny storyteller: how a computer learns to write, one letter at a time
results.csv
| step | train_loss | test_loss | train_first_guess_right | test_first_guess_right |
|---|---|---|---|---|
| 0 | 4.5353 | 4.5350 | 0.0115 | 0.0113 |
| 100 | 2.2940 | 2.3108 | 0.3270 | 0.3240 |
| 300 | 1.6979 | 1.7388 | 0.4849 | 0.4776 |
| 1000 | 1.2199 | 1.3766 | 0.6155 | 0.5788 |
| 2000 | 1.0776 | 1.3271 | 0.6572 | 0.5984 |