d400b921970fc0ddbd50d678945b35d5

This model is a fine-tuned version of google/mt5-xl on the Helsinki-NLP/opus_books [fr-nl] dataset. It achieves the following results on the evaluation set:

• Loss: 1.4014
• Data Size: 1.0
• Epoch Runtime: 461.9388
• Bleu: 11.4203

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Bleu
No log	0	0	8.2630	0	31.0323	0.0188
No log	1	1000	3.8062	0.0078	35.7253	0.2806
No log	2	2000	2.7161	0.0156	42.6108	0.7117
No log	3	3000	2.1331	0.0312	54.9550	5.7456
0.1162	4	4000	1.8301	0.0625	67.0948	6.6033
2.1822	5	5000	1.6886	0.125	93.9593	7.3803
0.1204	6	6000	1.5735	0.25	147.5408	8.3249
0.1529	7	7000	1.4545	0.5	251.6117	9.4436
1.4913	8.0	8000	1.3450	1.0	466.3258	10.3387
1.3101	9.0	9000	1.3107	1.0	464.3890	10.9153
1.1765	10.0	10000	1.2906	1.0	463.4392	11.0983
1.0573	11.0	11000	1.2905	1.0	462.1002	11.1935
0.9459	12.0	12000	1.3196	1.0	465.8401	11.4551
0.8613	13.0	13000	1.3363	1.0	463.7685	11.2788
0.7641	14.0	14000	1.3687	1.0	464.8621	11.3083
0.6781	15.0	15000	1.4014	1.0	461.9388	11.4203

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1