|
|
--- |
|
|
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_v3 |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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-7B-cgv1-ds_v3 |
|
|
|
|
|
This model is a fine-tuned version of [google/codegemma-7b](https://huggingface.co/google/codegemma-7b) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.0475 |
|
|
|
|
|
## 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: 12 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-------:|:-----:|:---------------:| |
|
|
| 0.7003 | 0.0530 | 50 | 0.6702 | |
|
|
| 0.5467 | 0.1061 | 100 | 0.5399 | |
|
|
| 0.4662 | 0.1591 | 150 | 0.4138 | |
|
|
| 0.3608 | 0.2121 | 200 | 0.3042 | |
|
|
| 0.3032 | 0.2652 | 250 | 0.2450 | |
|
|
| 0.2313 | 0.3182 | 300 | 0.2067 | |
|
|
| 0.1953 | 0.3713 | 350 | 0.1729 | |
|
|
| 0.1701 | 0.4243 | 400 | 0.1495 | |
|
|
| 0.1593 | 0.4773 | 450 | 0.1382 | |
|
|
| 0.1491 | 0.5304 | 500 | 0.1334 | |
|
|
| 0.1668 | 0.5834 | 550 | 0.1282 | |
|
|
| 0.1433 | 0.6364 | 600 | 0.1259 | |
|
|
| 0.1457 | 0.6895 | 650 | 0.1241 | |
|
|
| 0.1476 | 0.7425 | 700 | 0.1215 | |
|
|
| 0.139 | 0.7955 | 750 | 0.1176 | |
|
|
| 0.1209 | 0.8486 | 800 | 0.1159 | |
|
|
| 0.1365 | 0.9016 | 850 | 0.1148 | |
|
|
| 0.1239 | 0.9547 | 900 | 0.1157 | |
|
|
| 0.116 | 1.0077 | 950 | 0.1097 | |
|
|
| 0.1145 | 1.0607 | 1000 | 0.1104 | |
|
|
| 0.1187 | 1.1146 | 1050 | 0.1067 | |
|
|
| 0.117 | 1.1676 | 1100 | 0.1069 | |
|
|
| 0.1219 | 1.2206 | 1150 | 0.1059 | |
|
|
| 0.1192 | 1.2737 | 1200 | 0.1052 | |
|
|
| 0.1296 | 1.3267 | 1250 | 0.1023 | |
|
|
| 0.1016 | 1.3797 | 1300 | 0.1016 | |
|
|
| 0.1051 | 1.4328 | 1350 | 0.1011 | |
|
|
| 0.1207 | 1.4858 | 1400 | 0.1016 | |
|
|
| 0.1132 | 1.5388 | 1450 | 0.1031 | |
|
|
| 0.1143 | 1.5919 | 1500 | 0.0997 | |
|
|
| 0.1089 | 1.6449 | 1550 | 0.0988 | |
|
|
| 0.1164 | 1.6980 | 1600 | 0.0966 | |
|
|
| 0.1092 | 1.7510 | 1650 | 0.0961 | |
|
|
| 0.1056 | 1.8040 | 1700 | 0.0957 | |
|
|
| 0.1072 | 1.8571 | 1750 | 0.0948 | |
|
|
| 0.1029 | 1.9101 | 1800 | 0.0942 | |
|
|
| 0.1117 | 1.9631 | 1850 | 0.0931 | |
|
|
| 0.1126 | 2.0162 | 1900 | 0.0931 | |
|
|
| 0.104 | 2.0700 | 1950 | 0.0944 | |
|
|
| 0.1094 | 2.1230 | 2000 | 0.0925 | |
|
|
| 0.1044 | 2.1761 | 2050 | 0.0944 | |
|
|
| 0.0981 | 2.2291 | 2100 | 0.0926 | |
|
|
| 0.1031 | 2.2822 | 2150 | 0.0915 | |
|
|
| 0.0933 | 2.3352 | 2200 | 0.0919 | |
|
|
| 0.1085 | 2.3882 | 2250 | 0.0917 | |
|
|
| 0.1106 | 2.4413 | 2300 | 0.0905 | |
|
|
| 0.0988 | 2.4943 | 2350 | 0.0897 | |
|
|
| 0.0909 | 2.5473 | 2400 | 0.0883 | |
|
|
| 0.1025 | 2.6004 | 2450 | 0.0874 | |
|
|
| 0.1016 | 2.6534 | 2500 | 0.0873 | |
|
|
| 0.0927 | 2.7064 | 2550 | 0.0860 | |
|
|
| 0.0942 | 2.7595 | 2600 | 0.0854 | |
|
|
| 0.0888 | 2.8125 | 2650 | 0.0859 | |
|
|
| 0.091 | 2.8656 | 2700 | 0.0851 | |
|
|
| 0.0922 | 2.9186 | 2750 | 0.0855 | |
|
|
| 0.0949 | 2.9716 | 2800 | 0.0839 | |
|
|
| 0.0855 | 3.0247 | 2850 | 0.0841 | |
|
|
| 0.0955 | 3.0777 | 2900 | 0.0831 | |
|
|
| 0.0831 | 3.1307 | 2950 | 0.0817 | |
|
|
| 0.0843 | 3.1838 | 3000 | 0.0814 | |
|
|
| 0.0756 | 3.2368 | 3050 | 0.0812 | |
|
|
| 0.0893 | 3.2898 | 3100 | 0.0806 | |
|
|
| 0.0787 | 3.3429 | 3150 | 0.0827 | |
|
|
| 0.0842 | 3.3959 | 3200 | 0.0790 | |
|
|
| 0.079 | 3.4490 | 3250 | 0.0791 | |
|
|
| 0.0797 | 3.5020 | 3300 | 0.0773 | |
|
|
| 0.0774 | 3.5550 | 3350 | 0.0777 | |
|
|
| 0.0751 | 3.6081 | 3400 | 0.0779 | |
|
|
| 0.079 | 3.6611 | 3450 | 0.0781 | |
|
|
| 0.0849 | 3.7141 | 3500 | 0.0762 | |
|
|
| 0.0852 | 3.7672 | 3550 | 0.0759 | |
|
|
| 0.0742 | 3.8202 | 3600 | 0.0770 | |
|
|
| 0.0719 | 3.8732 | 3650 | 0.0755 | |
|
|
| 0.07 | 3.9263 | 3700 | 0.0757 | |
|
|
| 0.0778 | 3.9793 | 3750 | 0.0759 | |
|
|
| 0.0792 | 4.0324 | 3800 | 0.0751 | |
|
|
| 0.0705 | 4.0854 | 3850 | 0.0745 | |
|
|
| 0.0679 | 4.1384 | 3900 | 0.0741 | |
|
|
| 0.0619 | 4.1915 | 3950 | 0.0734 | |
|
|
| 0.0689 | 4.2445 | 4000 | 0.0731 | |
|
|
| 0.0653 | 4.2975 | 4050 | 0.0732 | |
|
|
| 0.0678 | 4.3506 | 4100 | 0.0733 | |
|
|
| 0.07 | 4.4036 | 4150 | 0.0719 | |
|
|
| 0.0656 | 4.4566 | 4200 | 0.0739 | |
|
|
| 0.062 | 4.5097 | 4250 | 0.0732 | |
|
|
| 0.0676 | 4.5627 | 4300 | 0.0718 | |
|
|
| 0.0668 | 4.6158 | 4350 | 0.0722 | |
|
|
| 0.0701 | 4.6688 | 4400 | 0.0718 | |
|
|
| 0.067 | 4.7218 | 4450 | 0.0709 | |
|
|
| 0.0686 | 4.7749 | 4500 | 0.0722 | |
|
|
| 0.0649 | 4.8279 | 4550 | 0.0751 | |
|
|
| 0.0711 | 4.8809 | 4600 | 0.0708 | |
|
|
| 0.0747 | 4.9340 | 4650 | 0.0711 | |
|
|
| 0.0622 | 4.9870 | 4700 | 0.0700 | |
|
|
| 0.0634 | 5.0400 | 4750 | 0.0695 | |
|
|
| 0.0714 | 5.0931 | 4800 | 0.0756 | |
|
|
| 0.0615 | 5.1461 | 4850 | 0.0732 | |
|
|
| 0.0612 | 5.1992 | 4900 | 0.0704 | |
|
|
| 0.0599 | 5.2522 | 4950 | 0.0686 | |
|
|
| 0.0567 | 5.3052 | 5000 | 0.0679 | |
|
|
| 0.0593 | 5.3583 | 5050 | 0.0673 | |
|
|
| 0.0576 | 5.4113 | 5100 | 0.0675 | |
|
|
| 0.0628 | 5.4643 | 5150 | 0.0664 | |
|
|
| 0.0572 | 5.5174 | 5200 | 0.0660 | |
|
|
| 0.06 | 5.5704 | 5250 | 0.0659 | |
|
|
| 0.0568 | 5.6234 | 5300 | 0.0660 | |
|
|
| 0.058 | 5.6765 | 5350 | 0.0656 | |
|
|
| 0.0559 | 5.7295 | 5400 | 0.0650 | |
|
|
| 0.0549 | 5.7826 | 5450 | 0.0652 | |
|
|
| 0.0605 | 5.8356 | 5500 | 0.0649 | |
|
|
| 0.0539 | 5.8886 | 5550 | 0.0641 | |
|
|
| 0.0567 | 5.9417 | 5600 | 0.0637 | |
|
|
| 0.0627 | 5.9971 | 5650 | 0.0633 | |
|
|
| 0.0576 | 6.0501 | 5700 | 0.0635 | |
|
|
| 0.0596 | 6.1032 | 5750 | 0.0654 | |
|
|
| 0.0751 | 6.1562 | 5800 | 0.0645 | |
|
|
| 0.0675 | 6.2092 | 5850 | 0.0636 | |
|
|
| 0.0575 | 6.2623 | 5900 | 0.0626 | |
|
|
| 0.0618 | 6.3153 | 5950 | 0.0626 | |
|
|
| 0.0641 | 6.3683 | 6000 | 0.0632 | |
|
|
| 0.0612 | 6.4214 | 6050 | 0.0616 | |
|
|
| 0.0599 | 6.4744 | 6100 | 0.0623 | |
|
|
| 0.0598 | 6.5274 | 6150 | 0.0607 | |
|
|
| 0.0597 | 6.5805 | 6200 | 0.0607 | |
|
|
| 0.0595 | 6.6335 | 6250 | 0.0602 | |
|
|
| 0.0612 | 6.6866 | 6300 | 0.0591 | |
|
|
| 0.058 | 6.7396 | 6350 | 0.0589 | |
|
|
| 0.0584 | 6.7926 | 6400 | 0.0580 | |
|
|
| 0.0544 | 6.8457 | 6450 | 0.0580 | |
|
|
| 0.0563 | 6.8987 | 6500 | 0.0576 | |
|
|
| 0.0569 | 6.9517 | 6550 | 0.0568 | |
|
|
| 0.0571 | 7.0048 | 6600 | 0.0572 | |
|
|
| 0.0463 | 7.0578 | 6650 | 0.0574 | |
|
|
| 0.0461 | 7.1108 | 6700 | 0.0570 | |
|
|
| 0.0468 | 7.1639 | 6750 | 0.0568 | |
|
|
| 0.051 | 7.2169 | 6800 | 0.0564 | |
|
|
| 0.0478 | 7.2700 | 6850 | 0.0561 | |
|
|
| 0.0487 | 7.3230 | 6900 | 0.0557 | |
|
|
| 0.0542 | 7.3760 | 6950 | 0.0563 | |
|
|
| 0.0504 | 7.4291 | 7000 | 0.0560 | |
|
|
| 0.046 | 7.4821 | 7050 | 0.0550 | |
|
|
| 0.0469 | 7.5351 | 7100 | 0.0554 | |
|
|
| 0.0473 | 7.5882 | 7150 | 0.0550 | |
|
|
| 0.0451 | 7.6412 | 7200 | 0.0548 | |
|
|
| 0.0519 | 7.6942 | 7250 | 0.0546 | |
|
|
| 0.0522 | 7.7473 | 7300 | 0.0543 | |
|
|
| 0.048 | 7.8003 | 7350 | 0.0546 | |
|
|
| 0.0519 | 7.8534 | 7400 | 0.0537 | |
|
|
| 0.0439 | 7.9064 | 7450 | 0.0537 | |
|
|
| 0.0474 | 7.9594 | 7500 | 0.0531 | |
|
|
| 0.0456 | 8.0125 | 7550 | 0.0533 | |
|
|
| 0.0439 | 8.0655 | 7600 | 0.0533 | |
|
|
| 0.0423 | 8.1185 | 7650 | 0.0535 | |
|
|
| 0.0405 | 8.1716 | 7700 | 0.0534 | |
|
|
| 0.0444 | 8.2246 | 7750 | 0.0539 | |
|
|
| 0.0416 | 8.2776 | 7800 | 0.0533 | |
|
|
| 0.0433 | 8.3307 | 7850 | 0.0541 | |
|
|
| 0.0466 | 8.3837 | 7900 | 0.0522 | |
|
|
| 0.047 | 8.4368 | 7950 | 0.0523 | |
|
|
| 0.0455 | 8.4898 | 8000 | 0.0528 | |
|
|
| 0.0471 | 8.5428 | 8050 | 0.0517 | |
|
|
| 0.042 | 8.5959 | 8100 | 0.0517 | |
|
|
| 0.0433 | 8.6489 | 8150 | 0.0520 | |
|
|
| 0.0488 | 8.7019 | 8200 | 0.0517 | |
|
|
| 0.0432 | 8.7550 | 8250 | 0.0521 | |
|
|
| 0.0472 | 8.8080 | 8300 | 0.0514 | |
|
|
| 0.042 | 8.8610 | 8350 | 0.0511 | |
|
|
| 0.0407 | 8.9141 | 8400 | 0.0505 | |
|
|
| 0.0415 | 8.9671 | 8450 | 0.0509 | |
|
|
| 0.038 | 9.0202 | 8500 | 0.0520 | |
|
|
| 0.0408 | 9.0732 | 8550 | 0.0521 | |
|
|
| 0.0367 | 9.1262 | 8600 | 0.0520 | |
|
|
| 0.0343 | 9.1793 | 8650 | 0.0507 | |
|
|
| 0.0379 | 9.2323 | 8700 | 0.0510 | |
|
|
| 0.0589 | 9.2853 | 8750 | 0.0554 | |
|
|
| 0.0398 | 9.3384 | 8800 | 0.0518 | |
|
|
| 0.04 | 9.3914 | 8850 | 0.0514 | |
|
|
| 0.0375 | 9.4444 | 8900 | 0.0521 | |
|
|
| 0.04 | 9.4975 | 8950 | 0.0503 | |
|
|
| 0.0381 | 9.5505 | 9000 | 0.0502 | |
|
|
| 0.0386 | 9.6036 | 9050 | 0.0495 | |
|
|
| 0.05 | 9.6566 | 9100 | 0.0519 | |
|
|
| 0.0389 | 9.7096 | 9150 | 0.0501 | |
|
|
| 0.0415 | 9.7627 | 9200 | 0.0499 | |
|
|
| 0.038 | 9.8157 | 9250 | 0.0503 | |
|
|
| 0.0433 | 9.8687 | 9300 | 0.0498 | |
|
|
| 0.036 | 9.9218 | 9350 | 0.0496 | |
|
|
| 0.0377 | 9.9748 | 9400 | 0.0488 | |
|
|
| 0.038 | 10.0278 | 9450 | 0.0495 | |
|
|
| 0.0384 | 10.0809 | 9500 | 0.0501 | |
|
|
| 0.035 | 10.1339 | 9550 | 0.0488 | |
|
|
| 0.0344 | 10.1870 | 9600 | 0.0484 | |
|
|
| 0.0356 | 10.2400 | 9650 | 0.0486 | |
|
|
| 0.0341 | 10.2930 | 9700 | 0.0501 | |
|
|
| 0.0333 | 10.3461 | 9750 | 0.0495 | |
|
|
| 0.0328 | 10.3991 | 9800 | 0.0496 | |
|
|
| 0.0337 | 10.4521 | 9850 | 0.0482 | |
|
|
| 0.0347 | 10.5052 | 9900 | 0.0489 | |
|
|
| 0.0318 | 10.5582 | 9950 | 0.0489 | |
|
|
| 0.0307 | 10.6112 | 10000 | 0.0481 | |
|
|
| 0.0344 | 10.6643 | 10050 | 0.0482 | |
|
|
| 0.0359 | 10.7173 | 10100 | 0.0490 | |
|
|
| 0.0325 | 10.7704 | 10150 | 0.0482 | |
|
|
| 0.0355 | 10.8234 | 10200 | 0.0495 | |
|
|
| 0.0361 | 10.8764 | 10250 | 0.0494 | |
|
|
| 0.0368 | 10.9295 | 10300 | 0.0486 | |
|
|
| 0.0378 | 10.9825 | 10350 | 0.0475 | |
|
|
| 0.0313 | 11.0355 | 10400 | 0.0475 | |
|
|
| 0.037 | 11.0886 | 10450 | 0.0473 | |
|
|
| 0.0377 | 11.1416 | 10500 | 0.0486 | |
|
|
| 0.0282 | 11.1946 | 10550 | 0.0479 | |
|
|
| 0.032 | 11.2477 | 10600 | 0.0498 | |
|
|
| 0.0387 | 11.3007 | 10650 | 0.0501 | |
|
|
| 0.0389 | 11.3538 | 10700 | 0.0486 | |
|
|
| 0.0333 | 11.4068 | 10750 | 0.0495 | |
|
|
| 0.032 | 11.4598 | 10800 | 0.0469 | |
|
|
| 0.0305 | 11.5129 | 10850 | 0.0479 | |
|
|
| 0.0362 | 11.5659 | 10900 | 0.0470 | |
|
|
| 0.0316 | 11.6189 | 10950 | 0.0487 | |
|
|
| 0.0337 | 11.6720 | 11000 | 0.0484 | |
|
|
| 0.0386 | 11.7250 | 11050 | 0.0479 | |
|
|
| 0.0313 | 11.7780 | 11100 | 0.0475 | |
|
|
| 0.0313 | 11.8311 | 11150 | 0.0466 | |
|
|
| 0.031 | 11.8841 | 11200 | 0.0474 | |
|
|
| 0.0318 | 11.9372 | 11250 | 0.0464 | |
|
|
| 0.0339 | 11.9902 | 11300 | 0.0475 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.12.0 |
|
|
- Transformers 4.44.2 |
|
|
- Pytorch 2.5.1+cu121 |
|
|
- Datasets 2.21.0 |
|
|
- Tokenizers 0.19.1 |
