code-bench-CodeGemma-7B-cgv1-ds
This model is a fine-tuned version of google/codegemma-7b on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0553
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
- train_batch_size: 1
- eval_batch_size: 3
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.2877 | 0.0530 | 50 | 1.7325 |
0.7107 | 0.1061 | 100 | 0.6972 |
0.5874 | 0.1591 | 150 | 0.5366 |
0.4688 | 0.2121 | 200 | 0.4286 |
0.386 | 0.2652 | 250 | 0.3401 |
0.2728 | 0.3182 | 300 | 0.2616 |
0.2257 | 0.3713 | 350 | 0.2191 |
0.1962 | 0.4243 | 400 | 0.1729 |
0.1726 | 0.4773 | 450 | 0.1531 |
0.1569 | 0.5304 | 500 | 0.1439 |
0.186 | 0.5834 | 550 | 0.1374 |
0.1467 | 0.6364 | 600 | 0.1326 |
0.1496 | 0.6895 | 650 | 0.1285 |
0.1484 | 0.7425 | 700 | 0.1265 |
0.1345 | 0.7955 | 750 | 0.1232 |
0.1321 | 0.8486 | 800 | 0.1199 |
0.1459 | 0.9016 | 850 | 0.1203 |
0.1305 | 0.9547 | 900 | 0.1174 |
0.1185 | 1.0077 | 950 | 0.1120 |
0.1208 | 1.0607 | 1000 | 0.1110 |
0.1112 | 1.1138 | 1050 | 0.1094 |
0.1154 | 1.1668 | 1100 | 0.1079 |
0.1131 | 1.2198 | 1150 | 0.1071 |
0.1192 | 1.2729 | 1200 | 0.1066 |
0.1217 | 1.3259 | 1250 | 0.1058 |
0.1055 | 1.3789 | 1300 | 0.1055 |
0.1032 | 1.4320 | 1350 | 0.1032 |
0.1133 | 1.4850 | 1400 | 0.1034 |
0.1142 | 1.5381 | 1450 | 0.1022 |
0.1091 | 1.5911 | 1500 | 0.1020 |
0.1024 | 1.6441 | 1550 | 0.0995 |
0.1166 | 1.6972 | 1600 | 0.1002 |
0.1057 | 1.7502 | 1650 | 0.0994 |
0.1056 | 1.8032 | 1700 | 0.0971 |
0.1059 | 1.8563 | 1750 | 0.0972 |
0.1063 | 1.9093 | 1800 | 0.0959 |
0.1025 | 1.9623 | 1850 | 0.0949 |
0.0938 | 2.0154 | 1900 | 0.0949 |
0.0973 | 2.0684 | 1950 | 0.0941 |
0.1037 | 2.1215 | 2000 | 0.0933 |
0.0888 | 2.1745 | 2050 | 0.0924 |
0.0877 | 2.2275 | 2100 | 0.0917 |
0.0874 | 2.2806 | 2150 | 0.0918 |
0.0874 | 2.3336 | 2200 | 0.0907 |
0.0959 | 2.3866 | 2250 | 0.0898 |
0.0954 | 2.4397 | 2300 | 0.0904 |
0.0887 | 2.4927 | 2350 | 0.0885 |
0.0827 | 2.5457 | 2400 | 0.0886 |
0.086 | 2.5988 | 2450 | 0.0869 |
0.0896 | 2.6518 | 2500 | 0.0861 |
0.0875 | 2.7049 | 2550 | 0.0862 |
0.0872 | 2.7579 | 2600 | 0.0863 |
0.0871 | 2.8109 | 2650 | 0.0850 |
0.0901 | 2.8640 | 2700 | 0.0842 |
0.078 | 2.9170 | 2750 | 0.0837 |
0.0878 | 2.9700 | 2800 | 0.0833 |
0.0807 | 3.0231 | 2850 | 0.0828 |
0.0846 | 3.0761 | 2900 | 0.0827 |
0.0741 | 3.1291 | 2950 | 0.0824 |
0.0778 | 3.1822 | 3000 | 0.0821 |
0.0755 | 3.2352 | 3050 | 0.0822 |
0.0728 | 3.2883 | 3100 | 0.0812 |
0.0758 | 3.3413 | 3150 | 0.0816 |
0.0779 | 3.3943 | 3200 | 0.0796 |
0.0705 | 3.4474 | 3250 | 0.0788 |
0.0725 | 3.5004 | 3300 | 0.0783 |
0.0707 | 3.5534 | 3350 | 0.0787 |
0.0659 | 3.6065 | 3400 | 0.0783 |
0.0698 | 3.6595 | 3450 | 0.0780 |
0.0702 | 3.7125 | 3500 | 0.0766 |
0.07 | 3.7656 | 3550 | 0.0768 |
0.0673 | 3.8186 | 3600 | 0.0760 |
0.0706 | 3.8717 | 3650 | 0.0751 |
0.0633 | 3.9247 | 3700 | 0.0741 |
0.0766 | 3.9777 | 3750 | 0.0740 |
0.0597 | 4.0308 | 3800 | 0.0741 |
0.0541 | 4.0838 | 3850 | 0.0742 |
0.0574 | 4.1368 | 3900 | 0.0734 |
0.0611 | 4.1899 | 3950 | 0.0727 |
0.0651 | 4.2429 | 4000 | 0.0723 |
0.0766 | 4.2991 | 4050 | 0.0699 |
0.0736 | 4.3522 | 4100 | 0.0696 |
0.0755 | 4.4052 | 4150 | 0.0695 |
0.0736 | 4.4582 | 4200 | 0.0692 |
0.0721 | 4.5113 | 4250 | 0.0686 |
0.071 | 4.5643 | 4300 | 0.0680 |
0.0675 | 4.6173 | 4350 | 0.0679 |
0.0714 | 4.6704 | 4400 | 0.0674 |
0.0648 | 4.7234 | 4450 | 0.0669 |
0.0729 | 4.7765 | 4500 | 0.0665 |
0.0656 | 4.8295 | 4550 | 0.0660 |
0.0707 | 4.8825 | 4600 | 0.0659 |
0.0703 | 4.9356 | 4650 | 0.0652 |
0.0669 | 4.9886 | 4700 | 0.0647 |
0.0665 | 5.0416 | 4750 | 0.0643 |
0.0573 | 5.0947 | 4800 | 0.0646 |
0.0628 | 5.1477 | 4850 | 0.0642 |
0.0574 | 5.2007 | 4900 | 0.0637 |
0.067 | 5.2538 | 4950 | 0.0632 |
0.06 | 5.3068 | 5000 | 0.0631 |
0.0637 | 5.3599 | 5050 | 0.0632 |
0.0602 | 5.4129 | 5100 | 0.0623 |
0.0592 | 5.4659 | 5150 | 0.0624 |
0.0567 | 5.5190 | 5200 | 0.0616 |
0.0634 | 5.5720 | 5250 | 0.0615 |
0.0577 | 5.6250 | 5300 | 0.0609 |
0.0534 | 5.6781 | 5350 | 0.0609 |
0.0555 | 5.7311 | 5400 | 0.0613 |
0.0601 | 5.7841 | 5450 | 0.0603 |
0.0533 | 5.8372 | 5500 | 0.0610 |
0.0598 | 5.8902 | 5550 | 0.0598 |
0.0603 | 5.9433 | 5600 | 0.0593 |
0.0606 | 5.9963 | 5650 | 0.0592 |
0.047 | 6.0493 | 5700 | 0.0596 |
0.0468 | 6.1024 | 5750 | 0.0589 |
0.0521 | 6.1554 | 5800 | 0.0586 |
0.045 | 6.2084 | 5850 | 0.0592 |
0.0504 | 6.2615 | 5900 | 0.0581 |
0.0474 | 6.3145 | 5950 | 0.0581 |
0.0497 | 6.3675 | 6000 | 0.0583 |
0.0519 | 6.4206 | 6050 | 0.0579 |
0.0467 | 6.4736 | 6100 | 0.0578 |
0.0475 | 6.5267 | 6150 | 0.0573 |
0.0506 | 6.5797 | 6200 | 0.0569 |
0.0486 | 6.6327 | 6250 | 0.0563 |
0.0441 | 6.6858 | 6300 | 0.0564 |
0.0528 | 6.7388 | 6350 | 0.0560 |
0.0506 | 6.7918 | 6400 | 0.0553 |
0.0491 | 6.8449 | 6450 | 0.0554 |
0.0458 | 6.8979 | 6500 | 0.0550 |
0.0458 | 6.9509 | 6550 | 0.0553 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.5.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 136
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
Model tree for Zacktree/code-bench-CodeGemma-7B-cgv1-ds
Base model
google/codegemma-7b