Hugging Face's logo

yalhessi
/
lemexp-task1-v2-template_full_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
Community
lemexp-task1-v2-template_full_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2 / README.md
yalhessi's picture
yalhessi
End of training
34bf20c verified
preview code
|
raw
history blame contribute delete
4.83 kB
metadata
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_full_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2
    results: []

lemexp-task1-v2-template_full_nodefs-deepseek-coder-1.3b-base-ddp-8lr-v2

This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1453

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
0.2767 0.2 3094 0.2776
0.2557 0.4 6188 0.2610
0.2507 0.6 9282 0.2532
0.2434 0.8 12376 0.2445
0.2395 1.0 15470 0.2353
0.2339 1.2 18564 0.2381
0.2332 1.4 21658 0.2260
0.2275 1.6 24752 0.2299
0.2298 1.8 27846 0.2205
0.2233 2.0 30940 0.2328
0.221 2.2 34034 0.2231
0.2201 2.4 37128 0.2136
0.2154 2.6 40222 0.2186
0.2152 2.8 43316 0.2148
0.2153 3.0 46410 0.2166
0.21 3.2 49504 0.2103
0.2094 3.4 52598 0.2103
0.2054 3.6 55692 0.2095
0.2046 3.8 58786 0.2053
0.2056 4.0 61880 0.2002
0.1989 4.2 64974 0.2069
0.1968 4.4 68068 0.1943
0.1948 4.6 71162 0.2035
0.1905 4.8 74256 0.1966
0.1909 5.0 77350 0.1933
0.1879 5.2 80444 0.1892
0.1877 5.4 83538 0.1933
0.186 5.6 86632 0.1895
0.1844 5.8 89726 0.1868
0.1815 6.0 92820 0.1869
0.1764 6.2 95914 0.1845
0.1764 6.4 99008 0.1874
0.1754 6.6 102102 0.1894
0.176 6.8 105196 0.1816
0.1724 7.0 108290 0.1799
0.1656 7.2 111384 0.1761
0.1637 7.4 114478 0.1751
0.1672 7.6 117572 0.1767
0.164 7.8 120666 0.1714
0.1637 8.0 123760 0.1714
0.1553 8.2 126854 0.1694
0.1538 8.4 129948 0.1700
0.1533 8.6 133042 0.1686
0.1561 8.8 136136 0.1641
0.1544 9.0 139230 0.1627
0.1457 9.2 142324 0.1582
0.1458 9.4 145418 0.1593
0.1447 9.6 148512 0.1590
0.1446 9.8 151606 0.1565
0.1405 10.0 154700 0.1557
0.1357 10.2 157794 0.1539
0.1338 10.4 160888 0.1528
0.1333 10.6 163982 0.1518
0.1319 10.8 167076 0.1509
0.1324 11.0 170170 0.1479
0.1264 11.2 173264 0.1487
0.1227 11.4 176358 0.1483
0.1234 11.6 179452 0.1458
0.1223 11.8 182546 0.1461
0.1226 12.0 185640 0.1453

Framework versions

  • PEFT 0.14.0
  • Transformers 4.47.0
  • Pytorch 2.5.1+cu124
  • Datasets 3.2.0
  • Tokenizers 0.21.0