mlpr-qwen2.5-0.5b-instruct-50ep-adaptive-v4
This model is a fine-tuned version of Qwen/Qwen2.5-0.5B-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.6362
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 12.3012 | 1.0 | 75 | 5.0283 |
| 8.5819 | 2.0 | 150 | 3.4116 |
| 7.072 | 3.0 | 225 | 3.3571 |
| 6.5538 | 4.0 | 300 | 3.9562 |
| 6.099 | 5.0 | 375 | 3.9585 |
| 5.4647 | 6.0 | 450 | 4.0752 |
| 4.7381 | 7.0 | 525 | 4.1101 |
| 4.3566 | 8.0 | 600 | 4.1768 |
| 3.796 | 9.0 | 675 | 4.2077 |
| 3.6005 | 10.0 | 750 | 4.2288 |
| 3.6009 | 11.0 | 825 | 4.2150 |
| 3.3501 | 12.0 | 900 | 4.2777 |
| 3.0221 | 13.0 | 975 | 4.2514 |
| 2.8183 | 14.0 | 1050 | 4.2798 |
| 2.8035 | 15.0 | 1125 | 4.2212 |
| 2.7384 | 16.0 | 1200 | 4.3070 |
| 2.5015 | 17.0 | 1275 | 4.3052 |
| 2.4855 | 18.0 | 1350 | 4.3162 |
| 2.3884 | 19.0 | 1425 | 4.3444 |
| 2.2479 | 20.0 | 1500 | 4.4013 |
| 2.1096 | 21.0 | 1575 | 4.3613 |
| 2.1082 | 22.0 | 1650 | 4.3914 |
| 1.9955 | 23.0 | 1725 | 4.3955 |
| 2.0264 | 24.0 | 1800 | 4.4025 |
| 2.0213 | 25.0 | 1875 | 4.4477 |
| 1.8517 | 26.0 | 1950 | 4.4073 |
| 1.7594 | 27.0 | 2025 | 4.4753 |
| 1.7804 | 28.0 | 2100 | 4.4606 |
| 1.7621 | 29.0 | 2175 | 4.4875 |
| 1.6965 | 30.0 | 2250 | 4.4962 |
| 1.5037 | 31.0 | 2325 | 4.4908 |
| 1.533 | 32.0 | 2400 | 4.5257 |
| 1.5308 | 33.0 | 2475 | 4.5147 |
| 1.6538 | 34.0 | 2550 | 4.4973 |
| 1.5472 | 35.0 | 2625 | 4.5542 |
| 1.4964 | 36.0 | 2700 | 4.5411 |
| 1.4629 | 37.0 | 2775 | 4.5561 |
| 1.452 | 38.0 | 2850 | 4.5826 |
| 1.4621 | 39.0 | 2925 | 4.5996 |
| 1.4902 | 40.0 | 3000 | 4.5710 |
| 1.4374 | 41.0 | 3075 | 4.6002 |
| 1.3841 | 42.0 | 3150 | 4.5988 |
| 1.2847 | 43.0 | 3225 | 4.6063 |
| 1.2819 | 44.0 | 3300 | 4.6166 |
| 1.2522 | 45.0 | 3375 | 4.6076 |
| 1.4089 | 46.0 | 3450 | 4.6172 |
| 1.2679 | 47.0 | 3525 | 4.6261 |
| 1.2993 | 48.0 | 3600 | 4.6130 |
| 1.3688 | 49.0 | 3675 | 4.6348 |
| 1.1921 | 50.0 | 3750 | 4.6362 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.5.1+cu124
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 240
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support