learning_rate: 0.0003
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
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
lr_scheduler_warmup_steps: 2
num_epochs: 3
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
3.2714	0.08	500	3.0344
3.4101	0.16	1000	2.9283
2.2733	0.24	1500	2.8800
1.8566	0.32	2000	2.8367
2.2548	0.4	2500	2.7989
3.447	0.48	3000	2.7752
2.304	0.56	3500	2.7567
3.1325	0.64	4000	2.7342
2.9546	0.72	4500	2.7158
2.9729	0.8	5000	2.7059
3.5754	0.88	5500	2.6906
3.8637	0.96	6000	2.6862
3.2746	1.04	6500	2.6655
2.9895	1.12	7000	2.6608
3.2698	1.2	7500	2.6436
2.1184	1.28	8000	2.6350
2.2134	1.3600	8500	2.6297
3.2429	1.44	9000	2.6204
2.8064	1.52	9500	2.6147
3.0127	1.6	10000	2.6057
3.2232	1.6800	10500	2.5991
2.2661	1.76	11000	2.5951
3.0668	1.8400	11500	2.5928
2.9571	1.92	12000	2.5847
3.2223	2.0	12500	2.5809
2.2106	2.08	13000	2.5771
3.1412	2.16	13500	2.5719
2.9079	2.24	14000	2.5671
2.5391	2.32	14500	2.5664
2.7341	2.4	15000	2.5613
3.0752	2.48	15500	2.5567
1.6035	2.56	16000	2.5563
2.6759	2.64	16500	2.5535
2.8205	2.7200	17000	2.5511
2.4317	2.8	17500	2.5474
2.816	2.88	18000	2.5473
3.0433	2.96	18500	2.5470

pratikshashetty5618/english-marathi-colloquial-translator

english-marathi-colloquial-translator