Benchmark
Link here.
| Step | Training Loss |
|---|---|
| 10 | 3.001900 |
| 20 | 2.650200 |
| 30 | 2.445400 |
| 40 | 2.266400 |
| 50 | 2.339200 |
| 60 | 2.192000 |
| 70 | 2.118200 |
| 80 | 2.128200 |
| 90 | 2.169100 |
| 100 | 2.079700 |
| 110 | 2.219000 |
| 120 | 2.111400 |
| 130 | 1.966000 |
| 140 | 2.141400 |
| 150 | 2.098100 |
| 160 | 1.998200 |
| 170 | 1.984500 |
| 180 | 2.098700 |
| 190 | 2.020800 |
| 200 | 1.996000 |
| 210 | 1.961200 |
| 220 | 2.020300 |
| 230 | 2.086700 |
| 240 | 1.967500 |
| 250 | 2.032500 |
| 260 | 1.967400 |
| 270 | 1.987300 |
| 280 | 1.944700 |
| 290 | 1.955300 |
| 300 | 1.908100 |
| 310 | 1.937300 |
| 320 | 1.863700 |
| 330 | 2.029000 |
| 340 | 1.776800 |
| 350 | 1.942100 |
| 360 | 1.876400 |
| 370 | 1.841400 |
| 380 | 1.885200 |
| 390 | 1.902400 |
| 400 | 1.893500 |
| 410 | 1.876300 |
| 420 | 1.913300 |
| 430 | 1.959800 |
| 440 | 1.869900 |
| 450 | 1.959500 |
| 460 | 1.824500 |
| 470 | 2.071600 |