learning_rate: 2e-05
train_batch_size: 8
eval_batch_size: 8
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: 0.05
num_epochs: 8
mixed_precision_training: Native AMP
label_smoothing_factor: 0.1

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
3.4446	1.0	1250	3.1152	0.3985	0.1795	0.2726	0.2723	80.006
2.9986	2.0	2500	3.0301	0.4112	0.1881	0.2785	0.2784	87.994
2.7535	3.0	3750	3.0240	0.4168	0.1944	0.291	0.2912	79.57
2.5724	4.0	5000	3.0551	0.42	0.195	0.2897	0.2899	91.921
2.4374	5.0	6250	3.0879	0.4216	0.1962	0.2936	0.2938	79.028
2.3269	6.0	7500	3.1304	0.4242	0.1964	0.2925	0.2926	87.542
2.2485	7.0	8750	3.1717	0.4219	0.1951	0.2916	0.2916	83.242

NastasiaM/mbart_cnn_summarization_model_10K_8ep_2e-5_batch_8_BEST_warmup

mbart_cnn_summarization_model_10K_8ep_2e-5_batch_8_BEST_warmup