learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 64
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Bleu Msl	Bleu Asl	Ter Msl	Ter Asl
No log	1.0	225	1.5057	0.0086	3.4312	40.2396	138.8368	49.3827
No log	2.0	450	0.9414	0.0086	43.1631	77.5441	43.8086	10.6428
1.758	3.0	675	0.5891	0.0086	61.0038	80.8049	28.1426	9.3231
1.758	4.0	900	0.4139	0.0086	70.0708	82.7822	21.4822	8.5568
0.6307	5.0	1125	0.3424	0.0086	73.0973	84.9287	19.5122	7.4500
0.6307	6.0	1350	0.3031	0.0086	77.0881	85.4456	16.1351	7.2371
0.3312	7.0	1575	0.2739	0.0086	79.0160	86.5112	14.9156	6.5986
0.3312	8.0	1800	0.2371	0.0086	79.2335	88.2693	13.9775	5.7897
0.2277	9.0	2025	0.2023	0.0086	79.0969	88.0941	13.5084	5.7471
0.2277	10.0	2250	0.1852	0.0086	81.0394	89.1417	13.0394	5.3640
0.2277	11.0	2475	0.1766	0.0086	81.7216	89.4270	12.1951	5.2363
0.1457	12.0	2700	0.1697	0.0086	81.9087	89.6539	12.2889	5.1511
0.1457	13.0	2925	0.1650	0.0086	81.8769	90.3167	12.0075	4.8106
0.1098	14.0	3150	0.1611	0.0086	82.2782	90.6362	11.6323	4.6403
0.1098	15.0	3375	0.1571	0.0086	81.9278	90.5687	11.3508	4.5977
0.0881	16.0	3600	0.1553	0.0086	83.3952	90.2462	10.7880	4.6828
0.0881	17.0	3825	0.1536	0.0086	83.1846	91.0621	10.8818	4.3423
0.0758	18.0	4050	0.1526	0.0086	82.4210	90.7751	11.2570	4.4700
0.0758	19.0	4275	0.1518	0.0086	82.6316	91.2367	10.7880	4.2997
0.065	20.0	4500	0.1499	0.0086	84.1579	91.7880	10.6004	4.0443
0.065	21.0	4725	0.1502	0.0086	83.1962	91.3530	10.3189	4.2571
0.065	22.0	4950	0.1500	0.0086	84.3437	91.5042	10.5066	4.1294
0.0577	23.0	5175	0.1495	0.0086	85.7608	91.8155	9.7561	4.0443
0.0577	24.0	5400	0.1490	0.0086	85.2304	91.9526	9.9437	3.9591
0.0533	25.0	5625	0.1491	0.0086	85.0375	91.4621	10.1313	4.1294
0.0533	26.0	5850	0.1482	0.0086	85.5992	91.4082	9.9437	4.1294
0.0489	27.0	6075	0.1484	0.0086	85.4254	91.6413	10.0375	4.0443
0.0489	28.0	6300	0.1483	0.0086	85.1178	91.6413	9.8499	4.0443
0.0481	29.0	6525	0.1482	0.0086	85.6291	91.3593	9.7561	4.2146
0.0481	30.0	6750	0.1481	0.0086	85.6291	91.2431	9.7561	4.2571

Framework versions

Transformers 4.47.0
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0

vania2911/9000

9000

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

vania2911/
9000