metadata
library_name: peft
license: gemma
base_model: google/codegemma-7b
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: code-bench-CodeGemma-7B-cgv1-ds_v2
results: []
code-bench-CodeGemma-7B-cgv1-ds_v2
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: 0.0737
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 |
|---|---|---|---|
| 1.5399 | 0.0530 | 50 | 1.1400 |
| 0.6618 | 0.1061 | 100 | 0.6524 |
| 0.5559 | 0.1591 | 150 | 0.5211 |
| 0.4396 | 0.2121 | 200 | 0.3873 |
| 0.3611 | 0.2652 | 250 | 0.3164 |
| 0.2679 | 0.3182 | 300 | 0.2416 |
| 0.2163 | 0.3713 | 350 | 0.1974 |
| 0.1916 | 0.4243 | 400 | 0.1638 |
| 0.1696 | 0.4773 | 450 | 0.1485 |
| 0.1601 | 0.5304 | 500 | 0.1416 |
| 0.1804 | 0.5834 | 550 | 0.1351 |
| 0.152 | 0.6364 | 600 | 0.1336 |
| 0.1532 | 0.6895 | 650 | 0.1317 |
| 0.1527 | 0.7425 | 700 | 0.1286 |
| 0.1446 | 0.7955 | 750 | 0.1250 |
| 0.1379 | 0.8486 | 800 | 0.1217 |
| 0.1451 | 0.9016 | 850 | 0.1227 |
| 0.1382 | 0.9547 | 900 | 0.1196 |
| 0.1218 | 1.0077 | 950 | 0.1170 |
| 0.1238 | 1.0607 | 1000 | 0.1174 |
| 0.1143 | 1.1138 | 1050 | 0.1171 |
| 0.1301 | 1.1668 | 1100 | 0.1136 |
| 0.1243 | 1.2198 | 1150 | 0.1123 |
| 0.1291 | 1.2729 | 1200 | 0.1119 |
| 0.1288 | 1.3259 | 1250 | 0.1104 |
| 0.1058 | 1.3789 | 1300 | 0.1082 |
| 0.1065 | 1.4320 | 1350 | 0.1059 |
| 0.1259 | 1.4850 | 1400 | 0.1068 |
| 0.1225 | 1.5381 | 1450 | 0.1051 |
| 0.1085 | 1.5911 | 1500 | 0.1035 |
| 0.1058 | 1.6441 | 1550 | 0.1031 |
| 0.1169 | 1.6972 | 1600 | 0.1022 |
| 0.1117 | 1.7502 | 1650 | 0.1010 |
| 0.1115 | 1.8032 | 1700 | 0.1008 |
| 0.1137 | 1.8563 | 1750 | 0.0995 |
| 0.1068 | 1.9093 | 1800 | 0.0990 |
| 0.1131 | 1.9623 | 1850 | 0.0979 |
| 0.1065 | 2.0154 | 1900 | 0.0974 |
| 0.0972 | 2.0684 | 1950 | 0.0966 |
| 0.1042 | 2.1215 | 2000 | 0.0955 |
| 0.0906 | 2.1745 | 2050 | 0.0948 |
| 0.0995 | 2.2275 | 2100 | 0.0949 |
| 0.0903 | 2.2806 | 2150 | 0.0939 |
| 0.0894 | 2.3336 | 2200 | 0.0938 |
| 0.0928 | 2.3866 | 2250 | 0.0925 |
| 0.1021 | 2.4397 | 2300 | 0.0922 |
| 0.0892 | 2.4927 | 2350 | 0.0911 |
| 0.0864 | 2.5457 | 2400 | 0.0901 |
| 0.0873 | 2.5988 | 2450 | 0.0895 |
| 0.0973 | 2.6518 | 2500 | 0.0887 |
| 0.0928 | 2.7049 | 2550 | 0.0883 |
| 0.0931 | 2.7579 | 2600 | 0.0883 |
| 0.0903 | 2.8109 | 2650 | 0.0871 |
| 0.0925 | 2.8640 | 2700 | 0.0867 |
| 0.0795 | 2.9170 | 2750 | 0.0854 |
| 0.0935 | 2.9700 | 2800 | 0.0851 |
| 0.0807 | 3.0231 | 2850 | 0.0860 |
| 0.0919 | 3.0761 | 2900 | 0.0846 |
| 0.0733 | 3.1291 | 2950 | 0.0840 |
| 0.0829 | 3.1822 | 3000 | 0.0840 |
| 0.0784 | 3.2352 | 3050 | 0.0833 |
| 0.0749 | 3.2883 | 3100 | 0.0827 |
| 0.0746 | 3.3413 | 3150 | 0.0828 |
| 0.0783 | 3.3943 | 3200 | 0.0824 |
| 0.0839 | 3.4474 | 3250 | 0.0815 |
| 0.0766 | 3.5004 | 3300 | 0.0810 |
| 0.0785 | 3.5534 | 3350 | 0.0804 |
| 0.0676 | 3.6065 | 3400 | 0.0800 |
| 0.0772 | 3.6595 | 3450 | 0.0796 |
| 0.0754 | 3.7125 | 3500 | 0.0794 |
| 0.0738 | 3.7656 | 3550 | 0.0790 |
| 0.0681 | 3.8186 | 3600 | 0.0788 |
| 0.0639 | 3.8717 | 3650 | 0.0788 |
| 0.069 | 3.9247 | 3700 | 0.0779 |
| 0.0725 | 3.9777 | 3750 | 0.0779 |
| 0.0692 | 4.0308 | 3800 | 0.0787 |
| 0.0597 | 4.0838 | 3850 | 0.0780 |
| 0.062 | 4.1368 | 3900 | 0.0775 |
| 0.0643 | 4.1899 | 3950 | 0.0774 |
| 0.0707 | 4.2429 | 4000 | 0.0766 |
| 0.0603 | 4.2959 | 4050 | 0.0771 |
| 0.0719 | 4.3490 | 4100 | 0.0767 |
| 0.0659 | 4.4020 | 4150 | 0.0769 |
| 0.0684 | 4.4551 | 4200 | 0.0764 |
| 0.0631 | 4.5081 | 4250 | 0.0768 |
| 0.0642 | 4.5611 | 4300 | 0.0766 |
| 0.0623 | 4.6142 | 4350 | 0.0766 |
| 0.0766 | 4.6672 | 4400 | 0.0765 |
| 0.0671 | 4.7202 | 4450 | 0.0764 |
| 0.0651 | 4.7733 | 4500 | 0.0762 |
| 0.0733 | 4.8295 | 4550 | 0.0750 |
| 0.0802 | 4.8825 | 4600 | 0.0749 |
| 0.0864 | 4.9356 | 4650 | 0.0748 |
| 0.0762 | 4.9886 | 4700 | 0.0747 |
| 0.0921 | 5.0416 | 4750 | 0.0747 |
| 0.0769 | 5.0947 | 4800 | 0.0747 |
| 0.0785 | 5.1477 | 4850 | 0.0746 |
| 0.0772 | 5.2007 | 4900 | 0.0745 |
| 0.0783 | 5.2538 | 4950 | 0.0745 |
| 0.0741 | 5.3068 | 5000 | 0.0745 |
| 0.08 | 5.3599 | 5050 | 0.0744 |
| 0.0813 | 5.4129 | 5100 | 0.0744 |
| 0.0764 | 5.4659 | 5150 | 0.0744 |
| 0.0752 | 5.5190 | 5200 | 0.0743 |
| 0.0778 | 5.5720 | 5250 | 0.0743 |
| 0.0813 | 5.6250 | 5300 | 0.0743 |
| 0.0701 | 5.6781 | 5350 | 0.0743 |
| 0.071 | 5.7311 | 5400 | 0.0742 |
| 0.0764 | 5.7841 | 5450 | 0.0742 |
| 0.0846 | 5.8372 | 5500 | 0.0742 |
| 0.0738 | 5.8902 | 5550 | 0.0742 |
| 0.0748 | 5.9433 | 5600 | 0.0741 |
| 0.0781 | 5.9963 | 5650 | 0.0741 |
| 0.0739 | 6.0493 | 5700 | 0.0741 |
| 0.069 | 6.1024 | 5750 | 0.0741 |
| 0.08 | 6.1554 | 5800 | 0.0741 |
| 0.0737 | 6.2084 | 5850 | 0.0740 |
| 0.075 | 6.2615 | 5900 | 0.0740 |
| 0.0752 | 6.3145 | 5950 | 0.0740 |
| 0.0859 | 6.3675 | 6000 | 0.0739 |
| 0.0872 | 6.4206 | 6050 | 0.0739 |
| 0.0768 | 6.4736 | 6100 | 0.0739 |
| 0.0742 | 6.5267 | 6150 | 0.0739 |
| 0.0779 | 6.5797 | 6200 | 0.0739 |
| 0.072 | 6.6327 | 6250 | 0.0739 |
| 0.0717 | 6.6858 | 6300 | 0.0738 |
| 0.0735 | 6.7388 | 6350 | 0.0738 |
| 0.0787 | 6.7918 | 6400 | 0.0738 |
| 0.0792 | 6.8449 | 6450 | 0.0738 |
| 0.0743 | 6.8979 | 6500 | 0.0737 |
| 0.074 | 6.9509 | 6550 | 0.0737 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.5.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1