metadata
library_name: peft
license: other
base_model: deepseek-ai/deepseek-coder-6.7b-base
tags:
- generated_from_trainer
model-index:
- name: lemexp-task1-v3-template_small_nodefs-deepseek-coder-6.7b-base
results: []
lemexp-task1-v3-template_small_nodefs-deepseek-coder-6.7b-base
This model is a fine-tuned version of deepseek-ai/deepseek-coder-6.7b-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1404
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: 0.0004
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 16
- total_eval_batch_size: 8
- 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 |
|---|---|---|---|
| 0.2947 | 0.2001 | 720 | 0.1929 |
| 0.1972 | 0.4001 | 1440 | 0.1784 |
| 0.1679 | 0.6002 | 2160 | 0.1608 |
| 0.1584 | 0.8002 | 2880 | 0.1579 |
| 0.1478 | 1.0003 | 3600 | 0.1473 |
| 0.1315 | 1.2003 | 4320 | 0.1439 |
| 0.128 | 1.4004 | 5040 | 0.1358 |
| 0.126 | 1.6004 | 5760 | 0.1329 |
| 0.1255 | 1.8005 | 6480 | 0.1322 |
| 0.1199 | 2.0006 | 7200 | 0.1269 |
| 0.108 | 2.2006 | 7920 | 0.1255 |
| 0.1047 | 2.4007 | 8640 | 0.1224 |
| 0.1038 | 2.6007 | 9360 | 0.1205 |
| 0.1043 | 2.8008 | 10080 | 0.1158 |
| 0.1042 | 3.0008 | 10800 | 0.1148 |
| 0.0884 | 3.2009 | 11520 | 0.1130 |
| 0.0903 | 3.4009 | 12240 | 0.1104 |
| 0.0889 | 3.6010 | 12960 | 0.1132 |
| 0.0885 | 3.8011 | 13680 | 0.1122 |
| 0.0882 | 4.0011 | 14400 | 0.1118 |
| 0.0753 | 4.2012 | 15120 | 0.1177 |
| 0.0764 | 4.4012 | 15840 | 0.1131 |
| 0.0755 | 4.6013 | 16560 | 0.1104 |
| 0.077 | 4.8013 | 17280 | 0.1090 |
| 0.0745 | 5.0014 | 18000 | 0.1091 |
| 0.0641 | 5.2014 | 18720 | 0.1118 |
| 0.0648 | 5.4015 | 19440 | 0.1087 |
| 0.0661 | 5.6016 | 20160 | 0.1093 |
| 0.0666 | 5.8016 | 20880 | 0.1060 |
| 0.065 | 6.0017 | 21600 | 0.1019 |
| 0.0565 | 6.2017 | 22320 | 0.1103 |
| 0.0575 | 6.4018 | 23040 | 0.1072 |
| 0.057 | 6.6018 | 23760 | 0.1105 |
| 0.057 | 6.8019 | 24480 | 0.1103 |
| 0.0566 | 7.0019 | 25200 | 0.1054 |
| 0.049 | 7.2020 | 25920 | 0.1135 |
| 0.0471 | 7.4021 | 26640 | 0.1102 |
| 0.0481 | 7.6021 | 27360 | 0.1120 |
| 0.0481 | 7.8022 | 28080 | 0.1071 |
| 0.0486 | 8.0022 | 28800 | 0.1090 |
| 0.0404 | 8.2023 | 29520 | 0.1156 |
| 0.0396 | 8.4023 | 30240 | 0.1120 |
| 0.0406 | 8.6024 | 30960 | 0.1184 |
| 0.0413 | 8.8024 | 31680 | 0.1123 |
| 0.0407 | 9.0025 | 32400 | 0.1140 |
| 0.033 | 9.2026 | 33120 | 0.1215 |
| 0.0345 | 9.4026 | 33840 | 0.1236 |
| 0.0344 | 9.6027 | 34560 | 0.1232 |
| 0.0351 | 9.8027 | 35280 | 0.1216 |
| 0.0336 | 10.0028 | 36000 | 0.1204 |
| 0.0289 | 10.2028 | 36720 | 0.1270 |
| 0.0286 | 10.4029 | 37440 | 0.1251 |
| 0.0292 | 10.6029 | 38160 | 0.1280 |
| 0.0287 | 10.8030 | 38880 | 0.1312 |
| 0.0297 | 11.0031 | 39600 | 0.1314 |
| 0.0256 | 11.2031 | 40320 | 0.1425 |
| 0.0254 | 11.4032 | 41040 | 0.1405 |
| 0.0254 | 11.6032 | 41760 | 0.1425 |
| 0.0255 | 11.8033 | 42480 | 0.1404 |
Framework versions
- PEFT 0.14.0
- Transformers 4.47.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.1