nllb-bt-sample-0

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: 3.4309
• Bleu: 0.9880

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
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
4.5077	2.5	500	3.7658	0.6291
3.5229	5.0	1000	3.4309	1.0214

Framework versions

• Transformers 4.33.0
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.13.3