metadata
license: mit
base_model: gpt2-large
tags:
- generated_from_trainer
datasets:
- scitldr
model-index:
- name: uplimit-project-3-gpt2-large
results: []
uplimit-project-3-gpt2-large
This model is a fine-tuned version of gpt2-large on the scitldr dataset. It achieves the following results on the evaluation set:
- Loss: 3.5296
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: 0.001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8474 | 0.4 | 800 | 2.7175 |
| 1.8784 | 0.8 | 1600 | 2.6618 |
| 1.6307 | 1.2 | 2400 | 2.7737 |
| 2.0121 | 1.61 | 3200 | 2.6673 |
| 2.033 | 2.01 | 4000 | 2.7799 |
| 1.8416 | 2.41 | 4800 | 2.8229 |
| 1.7133 | 2.81 | 5600 | 2.7827 |
| 1.681 | 3.21 | 6400 | 2.9556 |
| 1.5638 | 3.61 | 7200 | 2.9581 |
| 1.3838 | 4.02 | 8000 | 2.9749 |
| 1.2711 | 4.42 | 8800 | 2.9982 |
| 1.2488 | 4.82 | 9600 | 2.9858 |
| 0.9192 | 5.22 | 10400 | 3.1093 |
| 0.914 | 5.62 | 11200 | 3.1497 |
| 0.6613 | 6.02 | 12000 | 3.1170 |
| 0.6696 | 6.43 | 12800 | 3.1780 |
| 0.8281 | 6.83 | 13600 | 3.1630 |
| 0.3944 | 7.23 | 14400 | 3.3688 |
| 0.4512 | 7.63 | 15200 | 3.3493 |
| 0.313 | 8.03 | 16000 | 3.4182 |
| 0.3008 | 8.43 | 16800 | 3.4404 |
| 0.3054 | 8.84 | 17600 | 3.4577 |
| 0.2112 | 9.24 | 18400 | 3.5108 |
| 0.2435 | 9.64 | 19200 | 3.5296 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1