dense_swe_100m_mult_reseg_ep20_gemma
This model is a fine-tuned version of on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 5.6446
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
- 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 |
|---|---|---|---|
| 8.8597 | 0.7510 | 500 | 8.6242 |
| 7.4611 | 1.5017 | 1000 | 7.1015 |
| 6.3916 | 2.2523 | 1500 | 6.0770 |
| 5.6818 | 3.0030 | 2000 | 5.6144 |
| 5.3722 | 3.7540 | 2500 | 5.3655 |
| 5.0433 | 4.5047 | 3000 | 5.2152 |
| 4.9022 | 5.2554 | 3500 | 5.1209 |
| 4.7554 | 6.0060 | 4000 | 5.0575 |
| 4.5531 | 6.7570 | 4500 | 5.0187 |
| 4.3492 | 7.5077 | 5000 | 5.0189 |
| 4.2835 | 8.2584 | 5500 | 5.0368 |
| 4.2032 | 9.0090 | 6000 | 5.0475 |
| 4.0044 | 9.7600 | 6500 | 5.0640 |
| 3.8276 | 10.5107 | 7000 | 5.1151 |
| 3.7918 | 11.2614 | 7500 | 5.1775 |
| 3.7224 | 12.0120 | 8000 | 5.2051 |
| 3.5537 | 12.7630 | 8500 | 5.2526 |
| 3.4031 | 13.5137 | 9000 | 5.3180 |
| 3.3722 | 14.2644 | 9500 | 5.3804 |
| 3.3267 | 15.0150 | 10000 | 5.4193 |
| 3.1827 | 15.7661 | 10500 | 5.4642 |
| 3.07 | 16.5167 | 11000 | 5.5229 |
| 3.0515 | 17.2674 | 11500 | 5.5717 |
| 3.011 | 18.0180 | 12000 | 5.5973 |
| 2.9155 | 18.7691 | 12500 | 5.6225 |
| 2.8593 | 19.5197 | 13000 | 5.6441 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.1+cu128
- Datasets 3.6.0
- Tokenizers 0.22.1
- Downloads last month
- -