learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Bleu
1.1789	0.7057	500	1.1074	42.7039
1.1741	1.4107	1000	1.1058	42.7462
1.1837	2.1157	1500	1.1052	42.7927
1.1971	2.8215	2000	1.1035	42.8561
1.1862	3.5265	2500	1.1032	42.8571
1.179	4.2315	3000	1.1019	42.8509
1.1766	4.9372	3500	1.1014	42.8996
1.1867	5.6422	4000	1.1004	42.9536
1.1775	6.3472	4500	1.1006	42.9464
1.2007	7.0522	5000	1.0999	42.8951
1.192	7.7579	5500	1.0998	42.8856
1.1901	8.4629	6000	1.0997	42.9058
1.1722	9.1680	6500	1.0996	42.9150
1.1917	9.8737	7000	1.0996	42.9263

carlynamazed24/marian-mt-cb-en-lora-finetuned

marian-mt-cb-en-lora-finetuned