Nahuatl_Espanol_vn

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1464
• Bleu: 15.4218
• Gen Len: 45.5239

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	0.1064	100	1.1525	14.738	46.3599
No log	0.2128	200	1.1682	14.2823	45.9297
No log	0.3191	300	1.1739	14.2118	46.4243
No log	0.4255	400	1.1799	14.3198	45.9266
1.3984	0.5319	500	1.1771	14.0972	46.2179
1.3984	0.6383	600	1.1752	14.4083	45.8709
1.3984	0.7447	700	1.1756	14.1914	46.0949
1.3984	0.8511	800	1.1761	14.4131	46.0528
1.3984	0.9574	900	1.1727	14.1957	46.4856
1.3826	1.0638	1000	1.1768	14.7451	45.7873
1.3826	1.1702	1100	1.1727	14.6016	45.8654
1.3826	1.2766	1200	1.1726	14.6549	45.6857
1.3826	1.3830	1300	1.1693	14.586	45.6052
1.3826	1.4894	1400	1.1704	14.6483	45.6039
1.2932	1.5957	1500	1.1638	14.921	45.5508
1.2932	1.7021	1600	1.1649	14.7977	45.3693
1.2932	1.8085	1700	1.1580	14.9676	45.7072
1.2932	1.9149	1800	1.1567	14.794	45.5877
1.2932	2.0213	1900	1.1607	15.3066	45.677
1.2612	2.1277	2000	1.1569	15.1152	45.4122
1.2612	2.2340	2100	1.1553	15.2526	45.4026
1.2612	2.3404	2200	1.1521	15.2022	45.3518
1.2612	2.4468	2300	1.1505	15.3072	45.5873
1.2612	2.5532	2400	1.1500	15.417	45.5906
1.2095	2.6596	2500	1.1507	15.394	45.4383
1.2095	2.7660	2600	1.1501	15.4171	45.4846
1.2095	2.8723	2700	1.1472	15.4497	45.5049
1.2095	2.9787	2800	1.1464	15.4218	45.5239

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1