moe_e_het
This model is a fine-tuned version of on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 4.7569
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: 15297
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0 | 0 | 11.0250 |
| 7.0397 | 0.6537 | 1000 | 6.1960 |
| 6.3229 | 1.0 | 1530 | 5.4976 |
| 5.8474 | 1.3072 | 2000 | 5.2313 |
| 5.5294 | 1.9609 | 3000 | 4.9227 |
| 5.2675 | 2.6145 | 4000 | 4.7565 |
| 5.0232 | 3.2680 | 5000 | 4.6611 |
| 4.9874 | 3.9217 | 6000 | 4.5795 |
| 4.7947 | 4.5752 | 7000 | 4.5604 |
| 4.5741 | 5.2288 | 8000 | 4.5682 |
| 4.6031 | 5.8825 | 9000 | 4.5427 |
| 4.4044 | 6.5360 | 10000 | 4.5918 |
| 4.1874 | 7.1896 | 11000 | 4.6394 |
| 4.231 | 7.8432 | 12000 | 4.6484 |
| 4.2324 | 8.0 | 12240 | 4.6457 |
| 4.0585 | 8.4968 | 13000 | 4.7087 |
| 3.9213 | 9.1503 | 14000 | 4.7447 |
| 3.9334 | 9.8040 | 15000 | 4.7568 |
Framework versions
- Transformers 4.53.1
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1
- Downloads last month
- -