metadata
license: mit
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: openai-community/gpt2
model-index:
- name: outputs
results: []
outputs
This model is a fine-tuned version of openai-community/gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0494
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: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.03
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.9872 | 4.0 | 184 | 1.7517 |
| 1.769 | 8.0 | 368 | 1.3586 |
| 1.5071 | 12.0 | 552 | 1.2644 |
| 1.4096 | 16.0 | 736 | 1.2088 |
| 1.3405 | 20.0 | 920 | 1.1738 |
| 1.2971 | 24.0 | 1104 | 1.1478 |
| 1.2609 | 28.0 | 1288 | 1.1241 |
| 1.2328 | 32.0 | 1472 | 1.1112 |
| 1.2087 | 36.0 | 1656 | 1.0996 |
| 1.1919 | 40.0 | 1840 | 1.0918 |
| 1.1687 | 44.0 | 2024 | 1.0868 |
| 1.157 | 48.0 | 2208 | 1.0768 |
| 1.1436 | 52.0 | 2392 | 1.0724 |
| 1.1296 | 56.0 | 2576 | 1.0684 |
| 1.1147 | 60.0 | 2760 | 1.0657 |
| 1.1082 | 64.0 | 2944 | 1.0622 |
| 1.102 | 68.0 | 3128 | 1.0578 |
| 1.0925 | 72.0 | 3312 | 1.0573 |
| 1.0865 | 76.0 | 3496 | 1.0533 |
| 1.0786 | 80.0 | 3680 | 1.0534 |
| 1.076 | 84.0 | 3864 | 1.0523 |
| 1.0764 | 88.0 | 4048 | 1.0505 |
| 1.0685 | 92.0 | 4232 | 1.0495 |
| 1.0684 | 96.0 | 4416 | 1.0489 |
| 1.0662 | 100.0 | 4600 | 1.0494 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2