nllb-200-distilled-600M-finetuned-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: 0.7281
• Bleu: 63.3172
• Gen Len: 65.7

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
1.4803	1.0	695	0.9925	48.0036	68.092
1.0588	2.0	1390	0.8618	53.6714	67.794
0.8397	3.0	2085	0.8034	56.8749	67.316
0.7816	4.0	2780	0.7718	59.7588	65.822
0.7349	5.0	3475	0.7509	60.9155	66.205
0.6737	6.0	4170	0.7422	61.9048	65.348
0.6373	7.0	4865	0.7338	62.8549	65.607
0.617	8.0	5560	0.7308	63.6105	65.335
0.6068	9.0	6255	0.7276	63.452	65.594
0.5913	10.0	6950	0.7281	63.3172	65.7

Framework versions

• Transformers 4.31.0
• Pytorch 1.13.1
• Datasets 2.14.4
• Tokenizers 0.13.3