CoolFace
Modelpublic

yalhessi/lemexp-task1-v2-template_small-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos

sourceHugging Faceotherupdated 1y agoView on Hugging Face
0likes12downloads
Model Card

<!-- 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-deepseek-coder-1.3b-base-8lr-12epochs-normal-eos

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: 0.1544

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
  • —trainbatchsize: 2
  • —evalbatchsize: 2
  • —seed: 42
  • —distributed_type: multi-GPU
  • —num_devices: 8
  • —totaltrainbatch_size: 16
  • —totalevalbatch_size: 16
  • —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —num_epochs: 12
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
0.38950.20017200.2991
0.28970.400114400.2637
0.25360.600221600.2559
0.24530.800228800.2426
0.23351.000336000.2423
0.22141.200343200.2344
0.2171.400450400.2214
0.21541.600457600.2222
0.21331.800564800.2249
0.20952.000672000.2111
0.19842.200679200.2192
0.19732.400786400.2109
0.19382.600793600.1993
0.19412.8008100800.2004
0.19623.0008108000.1977
0.18133.2009115200.2014
0.18223.4009122400.2025
0.18143.6010129600.1900
0.17883.8011136800.1898
0.17954.0011144000.1856
0.16784.2012151200.1905
0.16934.4012158400.1903
0.16754.6013165600.1858
0.16814.8013172800.1844
0.16355.0014180000.1824
0.15375.2014187200.1821
0.1555.4015194400.1762
0.15595.6016201600.1837
0.15655.8016208800.1725
0.1516.0017216000.1710
0.14446.2017223200.1706
0.14476.4018230400.1719
0.14296.6018237600.1716
0.14286.8019244800.1673
0.14077.0019252000.1669
0.13357.2020259200.1650
0.1317.4021266400.1670
0.13147.6021273600.1617
0.12937.8022280800.1615
0.13098.0022288000.1615
0.11928.2023295200.1590
0.11868.4023302400.1567
0.11918.6024309600.1597
0.11818.8024316800.1533
0.11819.0025324000.1540
0.10689.2026331200.1538
0.10669.4026338400.1517
0.10719.6027345600.1535
0.1079.8027352800.1484
0.103810.0028360000.1516
0.095210.2028367200.1565
0.09410.4029374400.1533
0.094310.6029381600.1556
0.09410.8030388800.1504
0.094211.0031396000.1492
0.085611.2031403200.1560
0.08511.4032410400.1553
0.083211.6032417600.1560
0.083311.8033424800.1544

Framework versions

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