dense_swe_100m_mult_reseg_lr_div2
This model is a fine-tuned version of on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 5.1980
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use adamw_torch_fused 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: 1331
- training_steps: 13311
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 8.9334 | 0.7510 | 500 | 8.6299 |
| 7.6049 | 1.5017 | 1000 | 7.3508 |
| 6.8311 | 2.2523 | 1500 | 6.5959 |
| 6.2779 | 3.0030 | 2000 | 6.2161 |
| 6.0348 | 3.7540 | 2500 | 5.9760 |
| 5.7717 | 4.5047 | 3000 | 5.8110 |
| 5.646 | 5.2554 | 3500 | 5.6896 |
| 5.5173 | 6.0060 | 4000 | 5.6005 |
| 5.4179 | 6.7570 | 4500 | 5.5314 |
| 5.3151 | 7.5077 | 5000 | 5.4852 |
| 5.2745 | 8.2584 | 5500 | 5.4518 |
| 5.2335 | 9.0090 | 6000 | 5.4272 |
| 5.1822 | 9.7600 | 6500 | 5.4152 |
| 5.1971 | 10.5107 | 7000 | 5.4112 |
| 5.1597 | 11.2614 | 7500 | 5.3670 |
| 5.0918 | 12.0120 | 8000 | 5.3265 |
| 5.0162 | 12.7630 | 8500 | 5.2962 |
| 4.9366 | 13.5137 | 9000 | 5.2719 |
| 4.9049 | 14.2644 | 9500 | 5.2539 |
| 4.8697 | 15.0150 | 10000 | 5.2348 |
| 4.8049 | 15.7661 | 10500 | 5.2208 |
| 4.7515 | 16.5167 | 11000 | 5.2151 |
| 4.7336 | 17.2674 | 11500 | 5.2097 |
| 4.7154 | 18.0180 | 12000 | 5.2038 |
| 4.6696 | 18.7691 | 12500 | 5.2007 |
| 4.65 | 19.5197 | 13000 | 5.1996 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu128
- Datasets 3.6.0
- Tokenizers 0.22.1
- Downloads last month
- -