|
--- |
|
library_name: peft |
|
license: other |
|
base_model: deepseek-ai/deepseek-coder-1.3b-base |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: lemexp-task1-v2-template_small_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# lemexp-task1-v2-template_small_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2 |
|
|
|
This model is a fine-tuned version of [deepseek-ai/deepseek-coder-1.3b-base](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1494 |
|
|
|
## 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: 0.0008 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 8 |
|
- total_train_batch_size: 16 |
|
- total_eval_batch_size: 16 |
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- num_epochs: 12 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-------:|:-----:|:---------------:| |
|
| 0.4197 | 0.2002 | 721 | 0.2974 | |
|
| 0.3013 | 0.4003 | 1442 | 0.2783 | |
|
| 0.2634 | 0.6005 | 2163 | 0.2627 | |
|
| 0.2598 | 0.8007 | 2884 | 0.2501 | |
|
| 0.2469 | 1.0008 | 3605 | 0.2470 | |
|
| 0.2337 | 1.2010 | 4326 | 0.2299 | |
|
| 0.2303 | 1.4012 | 5047 | 0.2243 | |
|
| 0.2234 | 1.6013 | 5768 | 0.2270 | |
|
| 0.223 | 1.8015 | 6489 | 0.2233 | |
|
| 0.2184 | 2.0017 | 7210 | 0.2106 | |
|
| 0.2094 | 2.2018 | 7931 | 0.2106 | |
|
| 0.2075 | 2.4020 | 8652 | 0.2117 | |
|
| 0.2041 | 2.6022 | 9373 | 0.2049 | |
|
| 0.2031 | 2.8023 | 10094 | 0.2008 | |
|
| 0.2017 | 3.0025 | 10815 | 0.2097 | |
|
| 0.1897 | 3.2027 | 11536 | 0.1956 | |
|
| 0.1892 | 3.4028 | 12257 | 0.1954 | |
|
| 0.1868 | 3.6030 | 12978 | 0.1927 | |
|
| 0.1855 | 3.8032 | 13699 | 0.1917 | |
|
| 0.1862 | 4.0033 | 14420 | 0.1953 | |
|
| 0.1739 | 4.2035 | 15141 | 0.1908 | |
|
| 0.1749 | 4.4037 | 15862 | 0.1839 | |
|
| 0.174 | 4.6038 | 16583 | 0.1812 | |
|
| 0.1752 | 4.8040 | 17304 | 0.1810 | |
|
| 0.1725 | 5.0042 | 18025 | 0.1796 | |
|
| 0.16 | 5.2043 | 18746 | 0.1793 | |
|
| 0.162 | 5.4045 | 19467 | 0.1757 | |
|
| 0.1603 | 5.6047 | 20188 | 0.1768 | |
|
| 0.1607 | 5.8048 | 20909 | 0.1735 | |
|
| 0.1595 | 6.0050 | 21630 | 0.1759 | |
|
| 0.1509 | 6.2052 | 22351 | 0.1730 | |
|
| 0.1486 | 6.4053 | 23072 | 0.1712 | |
|
| 0.149 | 6.6055 | 23793 | 0.1668 | |
|
| 0.148 | 6.8057 | 24514 | 0.1635 | |
|
| 0.1475 | 7.0058 | 25235 | 0.1654 | |
|
| 0.1388 | 7.2060 | 25956 | 0.1637 | |
|
| 0.1372 | 7.4062 | 26677 | 0.1615 | |
|
| 0.136 | 7.6063 | 27398 | 0.1608 | |
|
| 0.1345 | 7.8065 | 28119 | 0.1591 | |
|
| 0.1343 | 8.0067 | 28840 | 0.1575 | |
|
| 0.1215 | 8.2068 | 29561 | 0.1554 | |
|
| 0.1236 | 8.4070 | 30282 | 0.1543 | |
|
| 0.1228 | 8.6072 | 31003 | 0.1560 | |
|
| 0.1244 | 8.8073 | 31724 | 0.1548 | |
|
| 0.1216 | 9.0075 | 32445 | 0.1563 | |
|
| 0.1093 | 9.2077 | 33166 | 0.1540 | |
|
| 0.109 | 9.4078 | 33887 | 0.1517 | |
|
| 0.11 | 9.6080 | 34608 | 0.1506 | |
|
| 0.1097 | 9.8082 | 35329 | 0.1503 | |
|
| 0.1098 | 10.0083 | 36050 | 0.1512 | |
|
| 0.0975 | 10.2085 | 36771 | 0.1527 | |
|
| 0.0966 | 10.4087 | 37492 | 0.1477 | |
|
| 0.0972 | 10.6088 | 38213 | 0.1456 | |
|
| 0.0971 | 10.8090 | 38934 | 0.1446 | |
|
| 0.0956 | 11.0092 | 39655 | 0.1484 | |
|
| 0.0887 | 11.2093 | 40376 | 0.1523 | |
|
| 0.0865 | 11.4095 | 41097 | 0.1496 | |
|
| 0.0874 | 11.6097 | 41818 | 0.1489 | |
|
| 0.0866 | 11.8098 | 42539 | 0.1494 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.14.0 |
|
- Transformers 4.47.0 |
|
- Pytorch 2.5.1+cu124 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.21.0 |