metadata
license: mit
base_model: gpt2
tags:
- code-completion
- 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: 1.9622
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: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.5115 | 0.15 | 200 | 4.8032 |
| 4.3012 | 0.3 | 400 | 3.6027 |
| 3.3162 | 0.45 | 600 | 2.8147 |
| 2.742 | 0.6 | 800 | 2.4321 |
| 2.4234 | 0.75 | 1000 | 2.2029 |
| 2.1911 | 0.9 | 1200 | 1.9622 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0a0+29c30b1
- Datasets 2.14.5
- Tokenizers 0.14.1