learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
num_epochs: 50

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rougel	Rougelsum
No log	0	0	2.8602	0	52.5614	0.0573	0.0324	0.0573	0.0573	0.0573
0.1379	1	17500	0.0908	0.0078	61.0524	0.9790	0.9790	0.9790	0.9790	0.9790
0.086	2	35000	0.0952	0.0156	71.1977	0.9801	0.9802	0.9801	0.9801	0.9801
0.0667	3	52500	0.0863	0.0312	90.2422	0.9796	0.9796	0.9796	0.9796	0.9797
0.0842	4	70000	0.0673	0.0625	127.7826	0.9838	0.9838	0.9838	0.9838	0.9839
0.0655	5	87500	0.0631	0.125	202.9388	0.9851	0.9851	0.9851	0.9852	0.9851
0.071	6	105000	0.0537	0.25	354.9037	0.9863	0.9863	0.9864	0.9863	0.9863
0.0003	7	122500	0.0457	0.5	644.2347	0.9885	0.9885	0.9885	0.9885	0.9885
0.0315	8.0	140000	0.0464	1.0	1242.5897	0.9885	0.9885	0.9885	0.9885	0.9885
0.0114	9.0	157500	0.0493	1.0	1248.1907	0.9889	0.9889	0.9889	0.9889	0.9889
0.0157	10.0	175000	0.0603	1.0	1239.6520	0.9872	0.9871	0.9872	0.9872	0.9872
0.0218	11.0	192500	0.0600	1.0	1239.5097	0.9890	0.9890	0.9890	0.9890	0.9890

contemmcm/a309576879fce3c35df54c51fa0ffe80

a309576879fce3c35df54c51fa0ffe80