learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
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: linear
num_epochs: 100

Training Loss	Epoch	Step	Validation Loss	Chrf++	Gen Len
41.4608	1.4706	4	2.4036	5.2353	188.3529
41.4608	2.9412	8	2.4034	5.2089	188.9412
41.4608	4.0	12	2.4040	5.6686	188.8824
41.4608	5.4706	16	2.4033	4.9653	188.9412
41.4608	6.9412	20	2.4033	4.6819	188.8235
41.4608	8.0	24	2.4026	4.4912	188.7647
41.4608	9.4706	28	2.4011	4.4052	188.7059
41.4608	10.9412	32	2.4008	4.5225	189.1765
41.4608	12.0	36	2.3990	5.1372	189.0588
41.4608	13.4706	40	2.3954	4.3585	189.0
41.4608	14.9412	44	2.3907	4.7137	188.2941
41.4608	16.0	48	2.3887	4.8434	188.9412
41.4608	17.4706	52	2.3836	4.7941	188.7647
41.4608	18.9412	56	2.3819	4.7571	189.0
41.4608	20.0	60	2.3742	5.3375	188.8235
41.4608	21.4706	64	2.3583	3.9694	186.2941
41.4608	22.9412	68	2.3528	3.7939	188.8235
41.4608	24.0	72	2.3437	4.7144	188.0
41.4608	25.4706	76	2.3437	4.1915	189.0588

Tsedeniya/finetuned_byT5large_pnb_joint

finetuned_byT5large_pnb_joint