mlm_old / README.md
psxjp5's picture
Update README.md
c300fde
|
raw
history blame
1.71 kB
metadata
license: mit
base_model: gpt2
tags:
  - generated_from_trainer
model-index:
  - name: mlm_new
    results: []

mlm_new

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 4.0711

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: 3e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Perplexity
12.948 0.99 22 5.1466 171.85
4.061 1.98 44 4.3114 74.54
3.7125 2.97 66 4.0807 59.19
3.6033 3.96 88 4.0553 57.70
3.5032 4.94 110 4.0514 57.48
3.4427 5.93 132 4.0879 59.61
3.3968 6.92 154 4.0711 58.62

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.2
  • Tokenizers 0.13.3