moe_g_medium_8exp
This model is a fine-tuned version of on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 4.5641
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 48952
- training_steps: 489524
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.6996 | 0.2043 | 10000 | 7.6592 |
| 6.2645 | 0.4086 | 20000 | 6.2143 |
| 5.4836 | 0.6128 | 30000 | 5.4589 |
| 5.1327 | 0.8171 | 40000 | 5.1078 |
| 4.8333 | 1.0214 | 50000 | 4.8943 |
| 4.7077 | 1.2257 | 60000 | 4.7191 |
| 4.6011 | 1.4299 | 70000 | 4.5950 |
| 4.5024 | 1.6342 | 80000 | 4.5029 |
| 4.4391 | 1.8385 | 90000 | 4.4284 |
| 4.1929 | 2.0428 | 100000 | 4.3720 |
| 4.1979 | 2.2471 | 110000 | 4.3348 |
| 4.1864 | 2.4513 | 120000 | 4.2980 |
| 4.1743 | 2.6556 | 130000 | 4.2615 |
| 4.1546 | 2.8599 | 140000 | 4.2289 |
| 3.8891 | 3.0642 | 150000 | 4.2212 |
| 3.9272 | 3.2684 | 160000 | 4.2077 |
| 3.9513 | 3.4727 | 170000 | 4.1873 |
| 3.9512 | 3.6770 | 180000 | 4.1657 |
| 3.9441 | 3.8813 | 190000 | 4.1458 |
| 3.6634 | 4.0856 | 200000 | 4.1753 |
| 3.7308 | 4.2898 | 210000 | 4.1710 |
| 3.7156 | 4.4941 | 220000 | 4.1554 |
| 3.7505 | 4.6984 | 230000 | 4.1376 |
| 3.7627 | 4.9027 | 240000 | 4.1210 |
| 3.4309 | 5.1069 | 250000 | 4.1895 |
| 3.4863 | 5.3112 | 260000 | 4.1895 |
| 3.5205 | 5.5155 | 270000 | 4.1772 |
| 3.5554 | 5.7198 | 280000 | 4.1610 |
| 3.5789 | 5.9241 | 290000 | 4.1443 |
| 3.2093 | 6.1283 | 300000 | 4.2485 |
| 3.2576 | 6.3326 | 310000 | 4.2520 |
| 3.3148 | 6.5369 | 320000 | 4.2403 |
| 3.3489 | 6.7412 | 330000 | 4.2251 |
| 3.3473 | 6.9454 | 340000 | 4.2094 |
| 3.0169 | 7.1497 | 350000 | 4.3366 |
| 3.0607 | 7.3540 | 360000 | 4.3474 |
| 3.1068 | 7.5583 | 370000 | 4.3378 |
| 3.116 | 7.7626 | 380000 | 4.3281 |
| 3.1245 | 7.9668 | 390000 | 4.3155 |
| 2.8132 | 8.1713 | 400000 | 4.4480 |
| 2.8409 | 8.3755 | 410000 | 4.4585 |
| 2.863 | 8.5798 | 420000 | 4.4558 |
| 2.891 | 8.7841 | 430000 | 4.4536 |
| 2.8756 | 8.9884 | 440000 | 4.4510 |
| 2.6193 | 9.1927 | 450000 | 4.5512 |
| 2.6376 | 9.3970 | 460000 | 4.5609 |
| 2.6431 | 9.6012 | 470000 | 4.5647 |
| 2.6512 | 9.8055 | 480000 | 4.5641 |
Framework versions
- Transformers 4.51.0
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1
- Downloads last month
- -