Edit model card

mbart-large-50-finetuned-sah-to-feat

This model is a fine-tuned version of facebook/mbart-large-50 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4514
  • Bleu: 5.5821
  • Gen Len: 199.0

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: 10

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
12.3174 1.0 24 6.7012 0.0148 199.0
5.2025 2.0 48 3.9396 0.0953 199.0
3.6977 3.0 72 2.1764 1.573 199.0
1.884 4.0 96 1.3887 3.5395 199.0
1.3753 5.0 120 0.9881 5.6111 199.0
0.9707 6.0 144 0.7452 5.2841 199.0
0.8008 7.0 168 0.6060 5.4831 199.0
0.6444 8.0 192 0.5174 5.4302 199.0
0.5689 9.0 216 0.4744 5.6898 199.0
0.5244 10.0 240 0.4514 5.5821 199.0

Framework versions

  • Transformers 4.30.2
  • Pytorch 2.0.1
  • Datasets 2.13.1
  • Tokenizers 0.13.3
Downloads last month
16
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.