Edit model card

t5-small-finetuned-eli5

This model is a fine-tuned version of t5-small on the eli5 dataset. It achieves the following results on the evaluation set:

  • Loss: 3.7275
  • Rouge1: 9.944
  • Rouge2: 1.908
  • Rougel: 8.0145
  • Rougelsum: 9.2275
  • Gen Len: 18.9988

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: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
3.9806 1.0 17040 3.7726 9.8475 1.872 7.9462 9.1258 18.9972
3.9458 2.0 34080 3.7369 9.9232 1.8981 7.9922 9.2061 18.9988
3.9355 3.0 51120 3.7275 9.944 1.908 8.0145 9.2275 18.9988

Framework versions

  • Transformers 4.22.2
  • Pytorch 1.12.1+cu113
  • Datasets 2.5.1
  • Tokenizers 0.12.1
Downloads last month
9

Dataset used to train Sandipan1994/t5-small-finetuned-eli5

Evaluation results