File size: 4,812 Bytes
ac68947 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 |
---
library_name: peft
license: other
base_model: deepseek-ai/deepseek-coder-1.3b-base
tags:
- generated_from_trainer
model-index:
- name: lemexp-task3-v2-small-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-task3-v2-small-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: 0.1304
## 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 |
|:-------------:|:-------:|:-----:|:---------------:|
| 0.382 | 0.2001 | 720 | 0.2739 |
| 0.2596 | 0.4001 | 1440 | 0.2392 |
| 0.2144 | 0.6002 | 2160 | 0.2247 |
| 0.2056 | 0.8002 | 2880 | 0.2076 |
| 0.1927 | 1.0003 | 3600 | 0.1997 |
| 0.1783 | 1.2003 | 4320 | 0.1963 |
| 0.1726 | 1.4004 | 5040 | 0.1855 |
| 0.1733 | 1.6004 | 5760 | 0.1810 |
| 0.1697 | 1.8005 | 6480 | 0.1782 |
| 0.1666 | 2.0006 | 7200 | 0.1705 |
| 0.1581 | 2.2006 | 7920 | 0.1725 |
| 0.1562 | 2.4007 | 8640 | 0.1662 |
| 0.1509 | 2.6007 | 9360 | 0.1702 |
| 0.1502 | 2.8008 | 10080 | 0.1607 |
| 0.1493 | 3.0008 | 10800 | 0.1635 |
| 0.1381 | 3.2009 | 11520 | 0.1604 |
| 0.139 | 3.4009 | 12240 | 0.1577 |
| 0.1388 | 3.6010 | 12960 | 0.1559 |
| 0.1353 | 3.8011 | 13680 | 0.1573 |
| 0.1375 | 4.0011 | 14400 | 0.1531 |
| 0.1231 | 4.2012 | 15120 | 0.1536 |
| 0.1277 | 4.4012 | 15840 | 0.1542 |
| 0.127 | 4.6013 | 16560 | 0.1529 |
| 0.1289 | 4.8013 | 17280 | 0.1467 |
| 0.1242 | 5.0014 | 18000 | 0.1439 |
| 0.1138 | 5.2014 | 18720 | 0.1419 |
| 0.1153 | 5.4015 | 19440 | 0.1431 |
| 0.1154 | 5.6016 | 20160 | 0.1398 |
| 0.1159 | 5.8016 | 20880 | 0.1434 |
| 0.1132 | 6.0017 | 21600 | 0.1394 |
| 0.1052 | 6.2017 | 22320 | 0.1361 |
| 0.1041 | 6.4018 | 23040 | 0.1401 |
| 0.1061 | 6.6018 | 23760 | 0.1355 |
| 0.105 | 6.8019 | 24480 | 0.1376 |
| 0.106 | 7.0019 | 25200 | 0.1322 |
| 0.0966 | 7.2020 | 25920 | 0.1357 |
| 0.0947 | 7.4021 | 26640 | 0.1322 |
| 0.0954 | 7.6021 | 27360 | 0.1341 |
| 0.0949 | 7.8022 | 28080 | 0.1329 |
| 0.097 | 8.0022 | 28800 | 0.1294 |
| 0.0857 | 8.2023 | 29520 | 0.1291 |
| 0.0852 | 8.4023 | 30240 | 0.1263 |
| 0.0859 | 8.6024 | 30960 | 0.1289 |
| 0.0843 | 8.8024 | 31680 | 0.1281 |
| 0.0853 | 9.0025 | 32400 | 0.1254 |
| 0.0755 | 9.2026 | 33120 | 0.1295 |
| 0.0753 | 9.4026 | 33840 | 0.1253 |
| 0.0776 | 9.6027 | 34560 | 0.1284 |
| 0.0762 | 9.8027 | 35280 | 0.1220 |
| 0.0757 | 10.0028 | 36000 | 0.1223 |
| 0.0679 | 10.2028 | 36720 | 0.1242 |
| 0.0687 | 10.4029 | 37440 | 0.1263 |
| 0.0678 | 10.6029 | 38160 | 0.1281 |
| 0.0678 | 10.8030 | 38880 | 0.1266 |
| 0.068 | 11.0031 | 39600 | 0.1266 |
| 0.0636 | 11.2031 | 40320 | 0.1287 |
| 0.0624 | 11.4032 | 41040 | 0.1280 |
| 0.0623 | 11.6032 | 41760 | 0.1303 |
| 0.0613 | 11.8033 | 42480 | 0.1304 |
### Framework versions
- PEFT 0.14.0
- Transformers 4.47.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0 |