| --- |
| license: mit |
| tags: |
| - generated_from_trainer |
| model-index: |
| - name: gpt2-synth |
| 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. --> |
|
|
| # gpt2-synth |
|
|
| This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.1783 |
|
|
| ## 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 | |
| |:-------------:|:-----:|:----:|:---------------:| |
| | 73.0212 | 0.02 | 10 | 65.7853 | |
| | 38.3819 | 0.04 | 20 | 20.2835 | |
| | 11.6035 | 0.07 | 30 | 9.2104 | |
| | 7.5582 | 0.09 | 40 | 3.6178 | |
| | 2.1601 | 0.11 | 50 | 1.5943 | |
| | 0.9769 | 0.13 | 60 | 0.9299 | |
| | 0.7577 | 0.15 | 70 | 0.6257 | |
| | 0.5434 | 0.18 | 80 | 0.4690 | |
| | 0.4756 | 0.2 | 90 | 0.5805 | |
| | 0.4247 | 0.22 | 100 | 0.3353 | |
| | 0.3179 | 0.24 | 110 | 0.2710 | |
| | 0.2476 | 0.26 | 120 | 0.2642 | |
| | 0.2604 | 0.29 | 130 | 0.2667 | |
| | 0.212 | 0.31 | 140 | 0.2723 | |
| | 0.2819 | 0.33 | 150 | 0.2356 | |
| | 0.2693 | 0.35 | 160 | 0.2283 | |
| | 0.2415 | 0.37 | 170 | 0.2297 | |
| | 0.2237 | 0.4 | 180 | 0.2207 | |
| | 0.2412 | 0.42 | 190 | 0.2204 | |
| | 0.2245 | 0.44 | 200 | 0.2191 | |
| | 0.2641 | 0.46 | 210 | 0.2112 | |
| | 0.22 | 0.48 | 220 | 0.2096 | |
| | 0.1912 | 0.51 | 230 | 0.2096 | |
| | 0.2152 | 0.53 | 240 | 0.2067 | |
| | 0.2497 | 0.55 | 250 | 0.2028 | |
| | 0.2042 | 0.57 | 260 | 0.2025 | |
| | 0.2263 | 0.59 | 270 | 0.2003 | |
| | 0.2047 | 0.62 | 280 | 0.1994 | |
| | 0.2081 | 0.64 | 290 | 0.1979 | |
| | 0.203 | 0.66 | 300 | 0.1954 | |
| | 0.1872 | 0.68 | 310 | 0.1954 | |
| | 0.1669 | 0.7 | 320 | 0.1958 | |
| | 0.1845 | 0.73 | 330 | 0.1935 | |
| | 0.1917 | 0.75 | 340 | 0.1913 | |
| | 0.2066 | 0.77 | 350 | 0.1880 | |
| | 0.1873 | 0.79 | 360 | 0.1880 | |
| | 0.1889 | 0.81 | 370 | 0.1867 | |
| | 0.2196 | 0.84 | 380 | 0.1853 | |
| | 0.2062 | 0.86 | 390 | 0.1848 | |
| | 0.1802 | 0.88 | 400 | 0.1836 | |
| | 0.1754 | 0.9 | 410 | 0.1828 | |
| | 0.1997 | 0.92 | 420 | 0.1813 | |
| | 0.2072 | 0.95 | 430 | 0.1790 | |
| | 0.19 | 0.97 | 440 | 0.1774 | |
| | 0.2208 | 0.99 | 450 | 0.1785 | |
| | 0.1706 | 1.01 | 460 | 0.1790 | |
| | 0.1534 | 1.03 | 470 | 0.1783 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.24.0 |
| - Pytorch 1.11.0+cu113 |
| - Datasets 2.6.1 |
| - Tokenizers 0.12.1 |
| |
