Downloads · 30 days
25
61% of all-time downloads
gpt2_tiny_baby_50M_32768_42
xiulinyang/gpt2_tiny_baby_50M_32768_42
gpt2_tiny_baby_50M_32768_42 is a machine learning model from xiulinyang. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
should probably proofread and complete it, then remove this comment. --
All-time downloads
41
Public
Repo size
3.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin46.7 MB · 96%
At a glance
- Model type
- gpt2
- Access
- Public
- Created
- Oct 12, 2025
- Updated
- Oct 12, 2025
- SHA
- 6442e76b
Model card
From the Hugging Face model README
<!-- 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. -->
gpt2_tiny_baby_50M_32768_42
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.7715
- Accuracy: 0.3332
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 40000
- training_steps: 100000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 8.283 | 0.43 | 2000 | 8.1606 | 0.1193 |
| 5.6979 | 0.87 | 4000 | 6.2471 | 0.1794 |
| 4.7397 | 1.3 | 6000 | 5.5927 | 0.2038 |
| 4.4648 | 1.73 | 8000 | 5.3240 | 0.2134 |
| 4.304 | 2.17 | 10000 | 5.1259 | 0.2224 |
| 4.1746 | 2.6 | 12000 | 4.9563 | 0.2346 |
| 4.0534 | 3.03 | 14000 | 4.8167 | 0.2452 |
| 3.9553 | 3.47 | 16000 | 4.6929 | 0.2529 |
| 3.8577 | 3.9 | 18000 | 4.5889 | 0.2604 |
| 3.7753 | 4.33 | 20000 | 4.5094 | 0.2662 |
| 3.6998 | 4.77 | 22000 | 4.4154 | 0.2721 |
| 3.6369 | 5.2 | 24000 | 4.3482 | 0.2773 |
| 3.5876 | 5.64 | 26000 | 4.2805 | 0.2822 |
| 3.5287 | 6.07 | 28000 | 4.2261 | 0.2864 |
| 3.4839 | 6.5 | 30000 | 4.1755 | 0.2907 |
| 3.4458 | 6.94 | 32000 | 4.1339 | 0.2953 |
| 3.4049 | 7.37 | 34000 | 4.1012 | 0.2991 |
| 3.369 | 7.8 | 36000 | 4.0649 | 0.3019 |
| 3.3284 | 8.24 | 38000 | 4.0342 | 0.3044 |
| 3.316 | 8.67 | 40000 | 4.0054 | 0.3075 |
| 3.2823 | 9.1 | 42000 | 3.9841 | 0.3102 |
| 3.258 | 9.54 | 44000 | 3.9635 | 0.3127 |
| 3.2452 | 9.97 | 46000 | 3.9409 | 0.3150 |
| 3.2171 | 10.4 | 48000 | 3.9224 | 0.3161 |
| 3.2129 | 10.84 | 50000 | 3.9127 | 0.3176 |
| 3.1892 | 11.27 | 52000 | 3.8949 | 0.3196 |
| 3.1881 | 11.7 | 54000 | 3.8819 | 0.3205 |
| 3.1666 | 12.14 | 56000 | 3.8752 | 0.3220 |
| 3.1579 | 12.57 | 58000 | 3.8623 | 0.3230 |
| 3.167 | 13.0 | 60000 | 3.8587 | 0.3239 |
| 3.1434 | 13.44 | 62000 | 3.8529 | 0.3233 |
| 3.1465 | 13.87 | 64000 | 3.8404 | 0.3256 |
| 3.1154 | 14.3 | 66000 | 3.8405 | 0.3253 |
| 3.1263 | 14.74 | 68000 | 3.8275 | 0.3269 |
| 3.1178 | 15.17 | 70000 | 3.8242 | 0.3271 |
| 3.1115 | 15.6 | 72000 | 3.8151 | 0.3280 |
| 3.1085 | 16.04 | 74000 | 3.8117 | 0.3287 |
| 3.0968 | 16.47 | 76000 | 3.8081 | 0.3288 |
| 3.1017 | 16.91 | 78000 | 3.8005 | 0.3298 |
| 3.0839 | 17.34 | 80000 | 3.8010 | 0.3296 |
| 3.0855 | 17.77 | 82000 | 3.7976 | 0.3305 |
| 3.0794 | 18.21 | 84000 | 3.7902 | 0.3313 |
| 3.0728 | 18.64 | 86000 | 3.7906 | 0.3309 |
| 3.0754 | 19.07 | 88000 | 3.7862 | 0.3313 |
| 3.0675 | 19.51 | 90000 | 3.7835 | 0.3321 |
| 3.0707 | 19.94 | 92000 | 3.7769 | 0.3330 |
| 3.0606 | 20.37 | 94000 | 3.7770 | 0.3324 |
| 3.0614 | 20.81 | 96000 | 3.7753 | 0.3327 |
| 3.0507 | 21.24 | 98000 | 3.7728 | 0.3330 |
| 3.0538 | 21.67 | 100000 | 3.7715 | 0.3332 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0+cu117
- Datasets 4.1.1
- Tokenizers 0.13.3
- Type
- gpt2
- Created
- Oct 12, 2025
- Updated
- Oct 12, 2025