| epoch/step,perplexity | |
| 0,51749.4883 | |
| 300,13.9451 | |
| 600,12.1401 | |
| 900,11.5001 | |
| 1200,11.1726 | |
| 1500,10.9386 | |
| 1800,10.7857 | |
| 2100,10.6209 | |
| 2400,10.5237 | |
| 2700,10.4321 | |
| 3000,10.4093 | |
| 3300,10.3519 | |
| 3600,10.2717 | |
| 3900,10.2833 | |
| 4200,10.2829 | |
| 4500,10.1701 | |
| 4800,10.1689 | |
| epoch/step,perplexity | |
| 0,51749.4883 | |
| 300,13.9451 | |
| 600,12.1401 | |
| 900,11.5001 | |
| 1200,11.1726 | |
| 1500,10.9386 | |
| 1800,10.7857 | |
| 2100,10.6209 | |
| 2400,10.5237 | |
| 2700,10.4321 | |
| 3000,10.4093 | |
| 3300,10.3519 | |
| 3600,10.2717 | |
| 3900,10.2833 | |
| 4200,10.2829 | |
| 4500,10.1701 | |
| 4800,10.1689 | |