|
|
--- |
|
|
library_name: transformers |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- accuracy |
|
|
model-index: |
|
|
- name: 100M__1208 |
|
|
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. --> |
|
|
|
|
|
# 100M__1208 |
|
|
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 3.4619 |
|
|
- Accuracy: 0.3767 |
|
|
|
|
|
## 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: 0.0006 |
|
|
- train_batch_size: 32 |
|
|
- eval_batch_size: 16 |
|
|
- seed: 1208 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 100 |
|
|
- num_epochs: 50 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Accuracy | Validation Loss | |
|
|
|:-------------:|:-------:|:-----:|:--------:|:---------------:| |
|
|
| 5.1151 | 0.1078 | 1000 | 0.2267 | 5.0271 | |
|
|
| 4.5792 | 0.2156 | 2000 | 0.2710 | 4.5047 | |
|
|
| 4.2988 | 0.3235 | 3000 | 0.2984 | 4.2390 | |
|
|
| 4.1617 | 0.4313 | 4000 | 0.3115 | 4.0934 | |
|
|
| 4.0655 | 0.5391 | 5000 | 0.3216 | 3.9903 | |
|
|
| 3.9934 | 0.6469 | 6000 | 0.3281 | 3.9243 | |
|
|
| 3.9432 | 0.7547 | 7000 | 0.3334 | 3.8671 | |
|
|
| 3.8869 | 0.8625 | 8000 | 0.3375 | 3.8181 | |
|
|
| 3.848 | 0.9704 | 9000 | 0.3415 | 3.7792 | |
|
|
| 3.7701 | 1.0782 | 10000 | 0.3447 | 3.7489 | |
|
|
| 3.7519 | 1.1860 | 11000 | 0.3467 | 3.7236 | |
|
|
| 3.7387 | 1.2938 | 12000 | 0.3488 | 3.6996 | |
|
|
| 3.7317 | 1.4016 | 13000 | 0.3512 | 3.6775 | |
|
|
| 3.6971 | 1.5094 | 14000 | 0.3538 | 3.6579 | |
|
|
| 3.6885 | 1.6173 | 15000 | 0.3551 | 3.6401 | |
|
|
| 3.6771 | 1.7251 | 16000 | 0.3570 | 3.6195 | |
|
|
| 3.6503 | 1.8329 | 17000 | 0.3589 | 3.6034 | |
|
|
| 3.6523 | 1.9407 | 18000 | 0.3600 | 3.5907 | |
|
|
| 3.5469 | 2.0485 | 19000 | 0.3615 | 3.5790 | |
|
|
| 3.553 | 2.1563 | 20000 | 0.3630 | 3.5688 | |
|
|
| 3.5623 | 2.2642 | 21000 | 0.3637 | 3.5586 | |
|
|
| 3.5464 | 2.3720 | 22000 | 0.3649 | 3.5480 | |
|
|
| 3.543 | 2.4798 | 23000 | 0.3659 | 3.5397 | |
|
|
| 3.5598 | 2.5876 | 24000 | 0.3667 | 3.5288 | |
|
|
| 3.533 | 2.6954 | 25000 | 0.3680 | 3.5185 | |
|
|
| 3.5272 | 2.8032 | 26000 | 0.3688 | 3.5118 | |
|
|
| 3.5211 | 2.9111 | 27000 | 0.3695 | 3.5022 | |
|
|
| 3.4424 | 3.0189 | 28000 | 0.3706 | 3.4983 | |
|
|
| 3.4427 | 3.1267 | 29000 | 0.3707 | 3.4975 | |
|
|
| 3.4427 | 3.2345 | 30000 | 0.3713 | 3.4904 | |
|
|
| 3.445 | 3.3423 | 31000 | 0.3723 | 3.4827 | |
|
|
| 3.4717 | 3.4501 | 32000 | 0.3730 | 3.4765 | |
|
|
| 3.4426 | 3.5580 | 33000 | 0.3736 | 3.4684 | |
|
|
| 3.4603 | 3.6658 | 34000 | 0.3741 | 3.4616 | |
|
|
| 3.4463 | 3.7736 | 35000 | 0.3749 | 3.4561 | |
|
|
| 3.4731 | 3.8814 | 36000 | 0.3756 | 3.4498 | |
|
|
| 3.4391 | 3.9892 | 37000 | 0.3763 | 3.4429 | |
|
|
| 3.3773 | 4.0970 | 38000 | 0.3766 | 3.4481 | |
|
|
| 3.3765 | 4.2049 | 39000 | 0.3770 | 3.4400 | |
|
|
| 3.3936 | 4.3127 | 40000 | 0.3779 | 3.4354 | |
|
|
| 3.3884 | 4.4205 | 41000 | 0.3781 | 3.4319 | |
|
|
| 3.3762 | 4.5283 | 42000 | 0.3790 | 3.4272 | |
|
|
| 3.375 | 4.6361 | 43000 | 0.3788 | 3.4225 | |
|
|
| 3.3747 | 4.7439 | 44000 | 0.3800 | 3.4165 | |
|
|
| 3.3985 | 4.8518 | 45000 | 0.3801 | 3.4101 | |
|
|
| 3.3853 | 4.9596 | 46000 | 0.3807 | 3.4061 | |
|
|
| 3.2829 | 5.0674 | 47000 | 0.3811 | 3.4088 | |
|
|
| 3.3263 | 5.1752 | 48000 | 0.3811 | 3.4104 | |
|
|
| 3.3098 | 5.2830 | 49000 | 0.3818 | 3.4050 | |
|
|
| 3.348 | 5.3908 | 50000 | 0.3819 | 3.4028 | |
|
|
| 3.331 | 5.4987 | 51000 | 0.3823 | 3.3962 | |
|
|
| 3.3412 | 5.6065 | 52000 | 0.3829 | 3.3931 | |
|
|
| 3.3414 | 5.7143 | 53000 | 0.3833 | 3.3883 | |
|
|
| 3.3438 | 5.8221 | 54000 | 0.3836 | 3.3822 | |
|
|
| 3.3077 | 5.9299 | 55000 | 0.3844 | 3.3785 | |
|
|
| 3.2654 | 6.0377 | 56000 | 0.3842 | 3.3808 | |
|
|
| 3.2674 | 6.1456 | 57000 | 0.3846 | 3.3812 | |
|
|
| 3.2691 | 6.2534 | 58000 | 0.3846 | 3.3798 | |
|
|
| 3.276 | 6.3612 | 59000 | 0.3851 | 3.3751 | |
|
|
| 3.2903 | 6.4690 | 60000 | 0.3858 | 3.3693 | |
|
|
| 3.2681 | 6.5768 | 61000 | 0.3861 | 3.3662 | |
|
|
| 3.2793 | 6.6846 | 62000 | 0.3863 | 3.3626 | |
|
|
| 3.2753 | 6.7925 | 63000 | 0.3866 | 3.3583 | |
|
|
| 3.2875 | 6.9003 | 64000 | 0.3873 | 3.3529 | |
|
|
| 3.2082 | 7.0081 | 65000 | 0.3875 | 3.3548 | |
|
|
| 3.2291 | 7.1159 | 66000 | 0.3872 | 3.3594 | |
|
|
| 3.2144 | 7.2237 | 67000 | 0.3872 | 3.3567 | |
|
|
| 3.2273 | 7.3315 | 68000 | 0.3884 | 3.3511 | |
|
|
| 3.2405 | 7.4394 | 69000 | 0.3886 | 3.3481 | |
|
|
| 3.2308 | 7.5472 | 70000 | 0.3887 | 3.3429 | |
|
|
| 3.2202 | 7.6550 | 71000 | 0.3893 | 3.3402 | |
|
|
| 3.2477 | 7.7628 | 72000 | 0.3899 | 3.3355 | |
|
|
| 3.2472 | 7.8706 | 73000 | 0.3900 | 3.3337 | |
|
|
| 3.2422 | 7.9784 | 74000 | 0.3905 | 3.3273 | |
|
|
| 3.1478 | 8.0863 | 75000 | 0.3902 | 3.3339 | |
|
|
| 3.1733 | 8.1941 | 76000 | 0.3906 | 3.3336 | |
|
|
| 3.1655 | 8.3019 | 77000 | 0.3909 | 3.3311 | |
|
|
| 3.16 | 8.4097 | 78000 | 0.3911 | 3.3274 | |
|
|
| 3.1795 | 8.5175 | 79000 | 0.3913 | 3.3251 | |
|
|
| 3.1806 | 8.6253 | 80000 | 0.3918 | 3.3206 | |
|
|
| 3.1844 | 8.7332 | 81000 | 0.3922 | 3.3171 | |
|
|
| 3.1883 | 8.8410 | 82000 | 0.3926 | 3.3136 | |
|
|
| 3.164 | 8.9488 | 83000 | 0.3929 | 3.3117 | |
|
|
| 3.1285 | 9.0566 | 84000 | 0.3927 | 3.3139 | |
|
|
| 3.127 | 9.1644 | 85000 | 0.3929 | 3.3135 | |
|
|
| 3.109 | 9.2722 | 86000 | 0.3933 | 3.3108 | |
|
|
| 3.1161 | 9.3801 | 87000 | 0.3934 | 3.3099 | |
|
|
| 3.1228 | 9.4879 | 88000 | 0.3937 | 3.3063 | |
|
|
| 3.1352 | 9.5957 | 89000 | 0.3940 | 3.3041 | |
|
|
| 3.127 | 9.7035 | 90000 | 0.3942 | 3.3023 | |
|
|
| 3.1309 | 9.8113 | 91000 | 0.3945 | 3.3003 | |
|
|
| 3.1271 | 9.9191 | 92000 | 0.3946 | 3.2983 | |
|
|
| 3.2875 | 10.0270 | 93000 | 3.4554 | 0.3782 | |
|
|
| 3.3636 | 10.1348 | 94000 | 3.4681 | 0.3759 | |
|
|
| 3.3804 | 10.2426 | 95000 | 3.4619 | 0.3767 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.47.0.dev0 |
|
|
- Pytorch 2.5.0+cu124 |
|
|
- Datasets 3.0.2 |
|
|
- Tokenizers 0.20.1 |
|
|
|
