|
--- |
|
license: mit |
|
base_model: gpt2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: mm2024-gpt-M-20240303T113807 |
|
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. --> |
|
|
|
# mm2024-gpt-M-20240303T113807 |
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1845 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 4096 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:------:|:---------------:| |
|
| 0.6928 | 0.24 | 15000 | 0.7607 | |
|
| 0.4003 | 0.48 | 30000 | 0.5600 | |
|
| 0.2811 | 0.72 | 45000 | 0.4320 | |
|
| 0.2085 | 0.96 | 60000 | 0.3571 | |
|
| 0.1548 | 1.2 | 75000 | 0.3006 | |
|
| 0.1241 | 1.44 | 90000 | 0.2771 | |
|
| 0.1056 | 1.68 | 105000 | 0.2569 | |
|
| 0.0933 | 1.92 | 120000 | 0.2407 | |
|
| 0.0843 | 2.16 | 135000 | 0.2245 | |
|
| 0.0788 | 2.4 | 150000 | 0.2189 | |
|
| 0.0746 | 2.64 | 165000 | 0.2034 | |
|
| 0.0706 | 2.88 | 180000 | 0.2050 | |
|
| 0.0668 | 3.12 | 195000 | 0.2051 | |
|
| 0.0647 | 3.36 | 210000 | 0.1980 | |
|
| 0.0628 | 3.6 | 225000 | 0.1930 | |
|
| 0.0611 | 3.84 | 240000 | 0.1865 | |
|
| 0.0585 | 4.08 | 255000 | 0.1887 | |
|
| 0.0573 | 4.32 | 270000 | 0.1857 | |
|
| 0.0561 | 4.56 | 285000 | 0.1849 | |
|
| 0.0554 | 4.8 | 300000 | 0.1845 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
