5e3859fc7436b50a645c682a5c544c91

This model is a fine-tuned version of google-t5/t5-3b on the Helsinki-NLP/opus_books [it-pt] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5490
• Data Size: 1.0
• Epoch Runtime: 31.1910
• Bleu: 7.1714

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	3.5600	0	1.5703	1.6985
No log	1	29	3.0980	0.0078	2.1857	2.7550
No log	2	58	2.7168	0.0156	6.9943	4.1219
No log	3	87	2.5005	0.0312	10.8658	4.9525
No log	4	116	2.2365	0.0625	16.7072	5.7843
No log	5	145	1.8806	0.125	21.6291	5.9050
0.2249	6	174	1.7636	0.25	27.2521	5.0542
0.2249	7	203	1.6438	0.5	23.4799	5.2144
0.2249	8.0	232	1.5360	1.0	32.2600	7.7956
1.074	9.0	261	1.4789	1.0	29.9934	7.7289
1.074	10.0	290	1.4629	1.0	31.8821	6.4226
1.2969	11.0	319	1.4596	1.0	35.6225	7.2500
1.2969	12.0	348	1.4604	1.0	24.3947	7.2294
1.0864	13.0	377	1.4706	1.0	26.7109	6.9801
0.9049	14.0	406	1.5123	1.0	29.0672	7.1244
0.9049	15.0	435	1.5490	1.0	31.1910	7.1714

Framework versions

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