learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_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: linear
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss
1.0878	0.1829	2500	0.8759
0.7323	0.3658	5000	0.6791
0.6637	0.5487	7500	0.6094
0.5856	0.7316	10000	0.5325
0.4145	0.9145	12500	0.1569
0.1222	1.0974	15000	0.0720
0.0607	1.2803	17500	0.0510
0.0521	1.4632	20000	0.0464
0.0481	1.6460	22500	0.0425
0.0449	1.8289	25000	0.0402
0.0431	2.0118	27500	0.0388
0.0528	2.1947	30000	0.0403
0.0385	2.3776	32500	0.0376
0.0378	2.5605	35000	0.0356
0.0368	2.7434	37500	0.0348
0.0356	2.9263	40000	0.0341
0.034	3.1092	42500	0.0335
0.0431	3.2921	45000	0.0345
0.0329	3.4750	47500	0.0326
0.0329	3.6579	50000	0.0322
0.0409	3.8408	52500	0.0330
0.0313	4.0237	55000	0.0313
0.0302	4.2066	57500	0.0310
0.0293	4.3895	60000	0.0306
0.0297	4.5724	62500	0.0303
0.0291	4.7553	65000	0.0300
0.0378	4.9382	67500	0.0310

bashyaldhiraj2067/new_epoch5_XPU_GEC_t5_char_nepali

new_epoch5_XPU_GEC_t5_char_nepali