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

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7727
• Bleu: 42.8708
• Gen Len: 13.335

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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.9294	1.0	1688	0.8364	39.0175	13.2637
0.7929	2.0	3376	0.7893	40.9994	13.303
0.7069	3.0	5064	0.7737	42.4125	13.292
0.6482	4.0	6752	0.7701	42.826	13.3287
0.6231	5.0	8440	0.7727	42.8708	13.335

Framework versions

• Transformers 4.29.0.dev0
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3