Team Ai
Modelpublic

zeetroid/code-bench-CodeGemma-7BIT-cg-nv9n_it_fs

sourceHugging Facegemmaupdated 1y agoView on Hugging Face
0likes20downloads
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. -->

code-bench-CodeGemma-7BIT-cg-nv9nitfs

This model is a fine-tuned version of google/codegemma-7b-it on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.0645

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: 5e-05
  • —trainbatchsize: 1
  • —evalbatchsize: 3
  • —seed: 42
  • —distributed_type: multi-GPU
  • —gradientaccumulationsteps: 8
  • —totaltrainbatch_size: 8
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_ratio: 0.03
  • —num_epochs: 5
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
0.74080.0530500.6774
0.51610.10611000.5082
0.43790.15911500.3828
0.3380.21212000.2834
0.26480.26522500.2229
0.20330.31823000.1773
0.18240.37133500.1469
0.15610.42434000.1352
0.14820.47734500.1283
0.13490.53045000.1212
0.15140.58345500.1157
0.13180.63646000.1137
0.13270.68956500.1119
0.13630.74257000.1109
0.12490.79557500.1075
0.11720.84868000.1067
0.11870.90168500.1077
0.11950.95479000.1049
0.10441.00779500.1022
0.11111.060710000.1025
0.10411.113810500.1019
0.10761.166811000.0989
0.10621.219811500.0991
0.11081.272912000.0968
0.10851.325912500.0961
0.09551.378913000.0949
0.09221.432013500.0943
0.10651.485014000.0935
0.10321.538114500.0920
0.0941.591115000.0910
0.0991.644115500.0903
0.0991.697216000.0895
0.09671.750216500.0893
0.09761.803217000.0887
0.09421.856317500.0876
0.09141.909318000.0865
0.09561.962318500.0855
0.08512.015419000.0853
0.08322.068419500.0851
0.0952.121520000.0854
0.07752.174520500.0840
0.08262.227521000.0828
0.07952.280621500.0831
0.08262.333622000.0828
0.08642.386622500.0810
0.08322.439723000.0802
0.08172.492723500.0796
0.07662.545724000.0789
0.08232.598824500.0783
0.07952.651825000.0780
0.07982.704925500.0771
0.08332.757926000.0770
0.07752.810926500.0760
0.08512.864027000.0755
0.06992.917027500.0746
0.08042.970028000.0743
0.06573.023128500.0746
0.07333.076129000.0735
0.0643.129129500.0733
0.06623.182230000.0731
0.06433.235230500.0722
0.06253.288331000.0721
0.06213.341331500.0718
0.06643.394332000.0716
0.06963.447432500.0708
0.06263.500433000.0705
0.06533.553433500.0701
0.05643.606534000.0697
0.06233.659534500.0692
0.06133.712535000.0688
0.06073.765635500.0686
0.05823.818636000.0685
0.05553.871736500.0680
0.05493.924737000.0677
0.06183.977737500.0673
0.05494.030838000.0674
0.05144.083838500.0674
0.04924.136839000.0670
0.05764.189939500.0670
0.05634.242940000.0665
0.05124.295940500.0665
0.05574.349041000.0663
0.06914.405241500.0663
0.06624.458242000.0661
0.0684.511342500.0659
0.06744.564343000.0657
0.06374.617343500.0655
0.07014.670444000.0654
0.06554.723444500.0651
0.06764.776545000.0650
0.06214.829545500.0649
0.06644.882546000.0647
0.06524.935646500.0646
0.06264.988647000.0645

Framework versions

  • —PEFT 0.12.0
  • —Transformers 4.44.2
  • —Pytorch 2.5.1+cu121
  • —Datasets 2.21.0
  • —Tokenizers 0.19.1