m2m100_418M-finetuned-sah-to-feat

This model is a fine-tuned version of facebook/m2m100_418M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0308
• Bleu: 4.6161
• Gen Len: 198.5197

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
No log	1.0	24	2.4936	1.8237	198.2756
No log	2.0	48	2.0218	3.342	198.8268
No log	3.0	72	1.7435	3.0434	198.874
No log	4.0	96	1.5399	3.8934	198.7953
No log	5.0	120	1.3805	3.5157	198.9685
No log	6.0	144	1.2383	4.2008	198.7559
No log	7.0	168	1.1430	4.1967	198.7244
No log	8.0	192	1.0837	3.9657	198.7874
No log	9.0	216	1.0501	4.0903	198.5354
No log	10.0	240	1.0308	4.6161	198.5197

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3