learning_rate: 1e-05
train_batch_size: 2
eval_batch_size: 2
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: 8
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.4798	1.0	500	1.8561	0.4043	0.1786	0.28	0.2792	85.69
1.5780	2.0	1000	1.6365	0.3989	0.172	0.278	0.2778	74.88
1.1267	3.0	1500	1.6541	0.3889	0.1698	0.2784	0.2781	66.04
1.0210	4.0	2000	1.8818	0.4058	0.1709	0.2855	0.2849	76.61
0.8366	5.0	2500	1.9876	0.3944	0.1643	0.2781	0.2773	67.75
0.7088	6.0	3000	2.0952	0.3944	0.1627	0.2682	0.2676	73.49
0.6115	7.0	3500	2.1958	0.3893	0.1554	0.2667	0.2659	71.66
0.5521	8.0	4000	2.2318	0.3849	0.1523	0.2633	0.2628	71.52

NastasiaM/mbart_cnn_summarization_model_1K_8ep_1e-5

mbart_cnn_summarization_model_1K_8ep_1e-5