metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: codeparrot-ds
results: []
codeparrot-ds
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0616
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5686 | 0.08 | 5000 | 1.7434 |
| 1.6823 | 0.15 | 10000 | 1.5212 |
| 1.5348 | 0.23 | 15000 | 1.4221 |
| 1.4546 | 0.31 | 20000 | 1.3565 |
| 1.3952 | 0.38 | 25000 | 1.3059 |
| 1.344 | 0.46 | 30000 | 1.2581 |
| 1.2978 | 0.54 | 35000 | 1.2138 |
| 1.2525 | 0.61 | 40000 | 1.1716 |
| 1.2109 | 0.69 | 45000 | 1.1324 |
| 1.1737 | 0.77 | 50000 | 1.0998 |
| 1.1441 | 0.84 | 55000 | 1.0769 |
| 1.1234 | 0.92 | 60000 | 1.0644 |
| 1.1155 | 1.0 | 65000 | 1.0616 |
Framework versions
- Transformers 4.29.2
- Pytorch 1.7.1
- Datasets 2.12.0
- Tokenizers 0.13.3