<!-- 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-no_ear-loto_muslim
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5300
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: 8
- eval_batch_size: 4
- seed: 21
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
73.4971 | 0.03 | 10 | 65.0370 |
31.3006 | 0.06 | 20 | 18.3819 |
11.2569 | 0.08 | 30 | 7.1196 |
3.7539 | 0.11 | 40 | 3.1582 |
1.8809 | 0.14 | 50 | 1.3415 |
1.0793 | 0.17 | 60 | 1.0276 |
0.905 | 0.2 | 70 | 0.7964 |
0.6987 | 0.23 | 80 | 0.7864 |
0.6243 | 0.25 | 90 | 0.7058 |
0.8346 | 0.28 | 100 | 0.6183 |
0.6377 | 0.31 | 110 | 0.6017 |
0.6088 | 0.34 | 120 | 0.5991 |
0.6074 | 0.37 | 130 | 0.5934 |
0.654 | 0.4 | 140 | 0.5844 |
0.6306 | 0.42 | 150 | 0.5815 |
0.6793 | 0.45 | 160 | 0.5951 |
0.6046 | 0.48 | 170 | 0.5731 |
0.6794 | 0.51 | 180 | 0.5730 |
0.6727 | 0.54 | 190 | 0.5682 |
0.5556 | 0.57 | 200 | 0.5611 |
0.6247 | 0.59 | 210 | 0.5570 |
0.6 | 0.62 | 220 | 0.5563 |
0.5464 | 0.65 | 230 | 0.5518 |
0.547 | 0.68 | 240 | 0.5516 |
0.4901 | 0.71 | 250 | 0.5468 |
0.5852 | 0.74 | 260 | 0.5472 |
0.5393 | 0.76 | 270 | 0.5459 |
0.5324 | 0.79 | 280 | 0.5468 |
0.5862 | 0.82 | 290 | 0.5438 |
0.5508 | 0.85 | 300 | 0.5404 |
0.6012 | 0.88 | 310 | 0.5368 |
0.525 | 0.91 | 320 | 0.5358 |
0.5764 | 0.93 | 330 | 0.5336 |
0.5239 | 0.96 | 340 | 0.5290 |
0.5902 | 0.99 | 350 | 0.5294 |
0.5059 | 1.02 | 360 | 0.5335 |
0.5069 | 1.05 | 370 | 0.5300 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.11.0+cu113
- Datasets 2.11.0
- Tokenizers 0.13.3