tiny-gpt2 / README.md
taufeeque's picture
Update to 1M parameter model
9763874
metadata
tags:
  - generated_from_trainer
datasets:
  - wikitext
metrics:
  - accuracy
model-index:
  - name: output_tiny
    results:
      - task:
          name: Causal Language Modeling
          type: text-generation
        dataset:
          name: wikitext wikitext-103-v1
          type: wikitext
          args: wikitext-103-v1
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.2132901596611274

output_tiny

This model is a fine-tuned version of gpt2_tiny_random on the wikitext wikitext-103-v1 dataset. It achieves the following results on the evaluation set:

  • Loss: 5.3359
  • Accuracy: 0.2133

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: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 50000

Training results

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.1
  • Datasets 2.8.0
  • Tokenizers 0.13.2