learning_rate: 5e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 0.1
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Bleu
No log	1.0	442	1.7653	0.0057	5.3892
2.8648	2.0	884	1.3633	0.0057	10.0945
1.5230	3.0	1326	1.2094	0.0057	13.0504
1.2053	4.0	1768	1.1409	0.0057	15.5247
1.0100	5.0	2210	1.0986	0.0057	16.3751
0.8717	6.0	2652	1.0945	0.0057	17.9580
0.7464	7.0	3094	1.0901	0.0057	18.4499
0.6584	8.0	3536	1.0888	0.0057	18.9487
0.6584	9.0	3978	1.0950	0.0057	19.4809
0.5903	10.0	4420	1.1023	0.0057	19.7404

danyanderson/m2m100-yemba-english-en2ybb

m2m100-yemba-english-en2ybb