nllb-200-distilled-600M-finetuned_augmented_MTbak_cleaned_ar-to-en

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9003
  • Bleu: 54.3021
  • Gen Len: 39.3817

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: 8
  • eval_batch_size: 8
  • 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
1.1718 1.0 2071 1.0140 46.727 39.6324
1.0097 2.0 4142 0.9450 50.6721 39.8258
0.8944 3.0 6213 0.9171 51.0894 39.9446
0.8192 4.0 8284 0.9010 52.8059 39.3494
0.7689 5.0 10355 0.8980 53.1204 39.5579
0.7511 6.0 12426 0.8946 53.5769 39.4814
0.685 7.0 14497 0.8968 53.8032 39.6365
0.6849 8.0 16568 0.8987 54.3402 39.5227
0.6548 9.0 18639 0.8992 54.4944 39.5619
0.6429 10.0 20710 0.9003 54.3021 39.3817

Framework versions

  • Transformers 4.33.0
  • Pytorch 1.13.1
  • Datasets 2.14.4
  • Tokenizers 0.13.3
Downloads last month
20
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.

Model tree for Jezia/nllb-200-distilled-600M-finetuned_augmented_MTbak_cleaned_ar-to-en

Finetuned
(85)
this model