m2m100_418M_fr_informal

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: 0.2971
• Bleu: 30.8965

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: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
4.2415	0.3554	500	0.3684	27.0259
0.3566	0.7107	1000	0.3164	28.8547
0.3309	1.0661	1500	0.3082	29.6687
0.3039	1.4215	2000	0.3049	29.8944
0.2966	1.7768	2500	0.3009	30.2278
0.2918	2.1322	3000	0.2994	30.5321
0.2741	2.4876	3500	0.2979	30.6031
0.2749	2.8429	4000	0.2971	30.8965

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0