qwen2_MetaMathQA_40K
This model is a fine-tuned version of Qwen/Qwen2-7B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1551
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: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 0.02
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.1756 | 0.0211 | 13 | 0.1624 |
0.1538 | 0.0421 | 26 | 0.1650 |
0.1596 | 0.0632 | 39 | 0.1662 |
0.1579 | 0.0842 | 52 | 0.1680 |
0.1577 | 0.1053 | 65 | 0.1701 |
0.1644 | 0.1264 | 78 | 0.1703 |
0.1691 | 0.1474 | 91 | 0.1722 |
0.1648 | 0.1685 | 104 | 0.1739 |
0.1707 | 0.1896 | 117 | 0.1749 |
0.1679 | 0.2106 | 130 | 0.1755 |
0.1715 | 0.2317 | 143 | 0.1761 |
0.1632 | 0.2527 | 156 | 0.1757 |
0.1706 | 0.2738 | 169 | 0.1760 |
0.1664 | 0.2949 | 182 | 0.1755 |
0.1713 | 0.3159 | 195 | 0.1743 |
0.1633 | 0.3370 | 208 | 0.1736 |
0.1652 | 0.3580 | 221 | 0.1737 |
0.1627 | 0.3791 | 234 | 0.1724 |
0.1646 | 0.4002 | 247 | 0.1723 |
0.1655 | 0.4212 | 260 | 0.1698 |
0.1727 | 0.4423 | 273 | 0.1692 |
0.1649 | 0.4633 | 286 | 0.1687 |
0.1715 | 0.4844 | 299 | 0.1687 |
0.1596 | 0.5055 | 312 | 0.1672 |
0.156 | 0.5265 | 325 | 0.1661 |
0.1577 | 0.5476 | 338 | 0.1655 |
0.1589 | 0.5687 | 351 | 0.1646 |
0.1621 | 0.5897 | 364 | 0.1635 |
0.1589 | 0.6108 | 377 | 0.1627 |
0.1514 | 0.6318 | 390 | 0.1616 |
0.1584 | 0.6529 | 403 | 0.1610 |
0.1592 | 0.6740 | 416 | 0.1599 |
0.1568 | 0.6950 | 429 | 0.1597 |
0.1584 | 0.7161 | 442 | 0.1587 |
0.1503 | 0.7371 | 455 | 0.1575 |
0.153 | 0.7582 | 468 | 0.1575 |
0.1549 | 0.7793 | 481 | 0.1569 |
0.1514 | 0.8003 | 494 | 0.1568 |
0.1485 | 0.8214 | 507 | 0.1564 |
0.1507 | 0.8424 | 520 | 0.1557 |
0.1536 | 0.8635 | 533 | 0.1557 |
0.1526 | 0.8846 | 546 | 0.1556 |
0.149 | 0.9056 | 559 | 0.1555 |
0.1461 | 0.9267 | 572 | 0.1552 |
0.1471 | 0.9478 | 585 | 0.1552 |
0.1484 | 0.9688 | 598 | 0.1552 |
0.1542 | 0.9899 | 611 | 0.1551 |
Framework versions
- PEFT 0.7.1
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for imdatta0/qwen2_MetaMathQA_40K
Base model
Qwen/Qwen2-7B