You can deploy this model and start earning money today!

contemmcm/
3edcb4ccf96181fb368f50434e0d0808

generated_from_trainer

Views

No views yet

3edcb4ccf96181fb368f50434e0d0808

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

Loss: 1.7370
Data Size: 1.0
Epoch Runtime: 185.2065
Bleu: 7.6635

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	7.9484	0	12.1540	0.0104
No log	1	367	3.4812	0.0078	13.8228	0.2724
No log	2	734	2.9004	0.0156	19.5233	0.5855
No log	3	1101	2.5128	0.0312	24.8288	0.7620
No log	4	1468	2.1384	0.0625	32.0654	1.3406
0.1626	5	1835	1.9743	0.125	44.2567	5.0637
2.3422	6	2202	1.9010	0.25	64.0586	5.6421
2.146	7	2569	1.8024	0.5	106.9209	6.0911
1.9187	8.0	2936	1.7193	1.0	188.3921	6.8771
1.7226	9.0	3303	1.6744	1.0	185.0243	7.2862
1.5496	10.0	3670	1.6637	1.0	188.0437	7.4163
1.4162	11.0	4037	1.6760	1.0	188.0105	7.4710
1.3277	12.0	4404	1.6833	1.0	185.4343	7.6708
1.2479	13.0	4771	1.6991	1.0	188.2047	7.6270
1.1243	14.0	5138	1.7370	1.0	185.2065	7.6635

Framework versions

Transformers 4.57.0
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

contemmcm/3edcb4ccf96181fb368f50434e0d0808

3edcb4ccf96181fb368f50434e0d0808

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

contemmcm/
3edcb4ccf96181fb368f50434e0d0808