codeparrot-ds / README.md
genchang's picture
End of training
0459213 verified
|
raw
history blame
1.87 kB
metadata
license: mit
base_model: gpt2
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: 5.8844

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: 10
  • num_epochs: 1
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
9.4802 0.09 10 8.0848
7.1562 0.18 20 7.0980
6.7304 0.27 30 6.8591
6.4473 0.36 40 6.5896
6.1701 0.45 50 6.3632
5.9031 0.54 60 6.1809
5.7632 0.64 70 6.0362
5.6136 0.73 80 5.9606
5.5312 0.82 90 5.9088
5.5063 0.91 100 5.8893
5.4845 1.0 110 5.8844

Framework versions

  • Transformers 4.36.0
  • Pytorch 2.0.1
  • Datasets 2.18.0
  • Tokenizers 0.15.2