| --- |
| tags: |
| - generated_from_trainer |
| datasets: |
| - generator |
| model-index: |
| - name: bert-concat |
| 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. --> |
|
|
| # bert-concat |
|
|
| This model is a fine-tuned version of [](https://huggingface.co/) on the generator dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 5.9507 |
|
|
| ## 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: 0.0005 |
| - train_batch_size: 64 |
| - eval_batch_size: 64 |
| - seed: 42 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: cosine |
| - lr_scheduler_warmup_steps: 1000 |
| - num_epochs: 14 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-----:|:-----:|:---------------:| |
| | 7.3397 | 0.25 | 500 | 6.6405 | |
| | 6.5835 | 0.51 | 1000 | 6.5183 | |
| | 6.4967 | 0.76 | 1500 | 6.4926 | |
| | 6.451 | 1.01 | 2000 | 6.4507 | |
| | 6.4104 | 1.26 | 2500 | 6.4097 | |
| | 6.3868 | 1.52 | 3000 | 6.4019 | |
| | 6.3717 | 1.77 | 3500 | 6.3789 | |
| | 6.3361 | 2.02 | 4000 | 6.3596 | |
| | 6.3099 | 2.28 | 4500 | 6.3345 | |
| | 6.2807 | 2.53 | 5000 | 6.3050 | |
| | 6.2578 | 2.78 | 5500 | 6.2843 | |
| | 6.2356 | 3.03 | 6000 | 6.2735 | |
| | 6.2017 | 3.29 | 6500 | 6.2527 | |
| | 6.1837 | 3.54 | 7000 | 6.2277 | |
| | 6.1682 | 3.79 | 7500 | 6.2102 | |
| | 6.1443 | 4.04 | 8000 | 6.1917 | |
| | 6.1128 | 4.3 | 8500 | 6.1767 | |
| | 6.1034 | 4.55 | 9000 | 6.1678 | |
| | 6.0838 | 4.8 | 9500 | 6.1552 | |
| | 6.0641 | 5.06 | 10000 | 6.1401 | |
| | 6.0417 | 5.31 | 10500 | 6.1350 | |
| | 6.0247 | 5.56 | 11000 | 6.1123 | |
| | 6.0125 | 5.81 | 11500 | 6.1082 | |
| | 6.0028 | 6.07 | 12000 | 6.1022 | |
| | 5.9788 | 6.32 | 12500 | 6.0895 | |
| | 5.9739 | 6.57 | 13000 | 6.0828 | |
| | 5.9545 | 6.83 | 13500 | 6.0687 | |
| | 5.9441 | 7.08 | 14000 | 6.0652 | |
| | 5.923 | 7.33 | 14500 | 6.0567 | |
| | 5.9115 | 7.58 | 15000 | 6.0492 | |
| | 5.9106 | 7.84 | 15500 | 6.0466 | |
| | 5.8943 | 8.09 | 16000 | 6.0315 | |
| | 5.8726 | 8.34 | 16500 | 6.0339 | |
| | 5.8665 | 8.59 | 17000 | 6.0243 | |
| | 5.8548 | 8.85 | 17500 | 6.0193 | |
| | 5.8431 | 9.1 | 18000 | 6.0111 | |
| | 5.8218 | 9.35 | 18500 | 6.0053 | |
| | 5.8193 | 9.61 | 19000 | 6.0026 | |
| | 5.8174 | 9.86 | 19500 | 5.9927 | |
| | 5.7954 | 10.11 | 20000 | 5.9873 | |
| | 5.7779 | 10.36 | 20500 | 5.9823 | |
| | 5.7749 | 10.62 | 21000 | 5.9799 | |
| | 5.7739 | 10.87 | 21500 | 5.9784 | |
| | 5.7582 | 11.12 | 22000 | 5.9757 | |
| | 5.7415 | 11.38 | 22500 | 5.9686 | |
| | 5.7467 | 11.63 | 23000 | 5.9650 | |
| | 5.7448 | 11.88 | 23500 | 5.9648 | |
| | 5.7372 | 12.13 | 24000 | 5.9585 | |
| | 5.7207 | 12.39 | 24500 | 5.9596 | |
| | 5.7264 | 12.64 | 25000 | 5.9546 | |
| | 5.7212 | 12.89 | 25500 | 5.9516 | |
| | 5.7142 | 13.14 | 26000 | 5.9553 | |
| | 5.7103 | 13.4 | 26500 | 5.9551 | |
| | 5.7093 | 13.65 | 27000 | 5.9527 | |
| | 5.7183 | 13.9 | 27500 | 5.9507 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.26.1 |
| - Pytorch 1.11.0+cu113 |
| - Datasets 2.13.0 |
| - Tokenizers 0.13.3 |
| |
