Edit model card

res_nw_lev_aragpt2-large

This model is a fine-tuned version of aubmindlab/aragpt2-large on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0510
  • Bleu: 0.1964
  • Rouge1: 0.5391
  • Rouge2: 0.3196
  • Rougel: 0.5372

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: 5e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 20.0

Training results

Training Loss Epoch Step Validation Loss Bleu Rouge1 Rouge2 Rougel
0.0936 1.0 10123 0.0602 0.0756 0.4154 0.1873 0.4125
0.0531 2.0 20246 0.0549 0.1086 0.4729 0.2428 0.4706
0.043 3.0 30369 0.0522 0.1457 0.5068 0.2846 0.5043
0.0347 4.0 40492 0.0510 0.1964 0.5391 0.3196 0.5372
0.029 5.0 50615 0.0511 0.2436 0.5627 0.3532 0.5608
0.0247 6.0 60738 0.0522 0.2849 0.5778 0.3798 0.5761
0.0218 7.0 70861 0.0531 0.3156 0.5943 0.4031 0.5925
0.0195 8.0 80984 0.0551 0.3369 0.5990 0.4135 0.5972
0.0179 9.0 91107 0.0563 0.3488 0.6067 0.4231 0.6048

Framework versions

  • Transformers 4.45.0.dev0
  • Pytorch 2.3.1+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1
Downloads last month
2
Safetensors
Model size
792M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for nlparabic/res_nw_lev_aragpt2-large

Finetuned
(5)
this model