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
No log	0	0	1.7888	0	1.1379	0.1646	0.0958
No log	1	170	1.6679	0.0078	1.5366	0.2792	0.0755
No log	2	340	1.6111	0.0156	2.0178	0.3292	0.1437
No log	3	510	1.3621	0.0312	2.8148	0.5417	0.3783
No log	4	680	0.7766	0.0625	3.5131	0.8458	0.7089
0.0647	5	850	0.3235	0.125	4.5176	0.9021	0.7570
0.0647	6	1020	0.2060	0.25	6.6535	0.9333	0.7850
0.25	7	1190	0.2341	0.5	10.3265	0.9396	0.9293
0.1677	8.0	1360	0.1876	1.0	18.4115	0.9583	0.9627
0.1115	9.0	1530	0.2414	1.0	17.7747	0.9417	0.9511
0.0714	10.0	1700	0.1578	1.0	17.8129	0.975	0.9771
0.0423	11.0	1870	0.1657	1.0	18.0082	0.9708	0.9747
0.0393	12.0	2040	0.1994	1.0	18.6068	0.9667	0.9712
0.0156	13.0	2210	0.2343	1.0	19.0090	0.9667	0.9703
0.016	14.0	2380	0.2414	1.0	19.5014	0.9646	0.9486

contemmcm/fd192deec337503653744dbe822995e2

fd192deec337503653744dbe822995e2