moe_d_hom
This model is a fine-tuned version of on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 5.1610
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: 32
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 16
- total_train_batch_size: 1024
- total_eval_batch_size: 64
- 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: 500
- training_steps: 20871
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0 | 0 | 10.9662 |
| 7.1285 | 0.4791 | 1000 | 6.8667 |
| 5.8595 | 0.9582 | 2000 | 5.7555 |
| 5.4049 | 1.4370 | 3000 | 5.3745 |
| 5.1883 | 1.9161 | 4000 | 5.1555 |
| 4.9343 | 2.3948 | 5000 | 5.0342 |
| 4.8581 | 2.8739 | 6000 | 4.9327 |
| 4.6274 | 3.3526 | 7000 | 4.8933 |
| 4.6081 | 3.8317 | 8000 | 4.8336 |
| 4.3692 | 4.3105 | 9000 | 4.8467 |
| 4.3805 | 4.7896 | 10000 | 4.8122 |
| 4.1175 | 5.2683 | 11000 | 4.8594 |
| 4.1546 | 5.7474 | 12000 | 4.8420 |
| 3.8785 | 6.2261 | 13000 | 4.9100 |
| 3.9278 | 6.7053 | 14000 | 4.9118 |
| 3.6634 | 7.1840 | 15000 | 4.9874 |
| 3.716 | 7.6631 | 16000 | 5.0023 |
| 3.4779 | 8.1418 | 17000 | 5.0715 |
| 3.528 | 8.6209 | 18000 | 5.0938 |
| 3.5338 | 9.0 | 18792 | 5.0963 |
| 3.3383 | 9.0997 | 19000 | 5.1385 |
| 3.3624 | 9.5788 | 20000 | 5.1584 |
Framework versions
- Transformers 4.53.1
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1
- Downloads last month
- -