learning_rate: 3e-05
train_batch_size: 32
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
num_epochs: 15
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Bleu
No log	1.0	33	0.1666	5.4192
No log	2.0	66	0.1532	15.0023
No log	3.0	99	0.1466	14.3780
No log	4.0	132	0.1448	15.1544
No log	5.0	165	0.1433	15.4036
No log	6.0	198	0.1425	14.6474
No log	7.0	231	0.1418	14.1405
No log	8.0	264	0.1417	20.3918
No log	9.0	297	0.1420	14.2228
No log	10.0	330	0.1415	20.6111
No log	11.0	363	0.1418	19.4349
No log	12.0	396	0.1417	19.0929
No log	13.0	429	0.1419	19.9174
No log	14.0	462	0.1419	19.9016
No log	15.0	495	0.1418	19.8220

msmaje/hausaenglishchatbotmodel

hausaenglishchatbotmodel