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
61.2327	1.0	2	2.5877	3.1494	189.25
61.2327	2.0	4	2.5882	3.1847	189.2905
61.2327	3.0	6	2.5884	3.0828	189.2365
61.2327	4.0	8	2.5884	2.9804	189.1486
61.2327	5.0	10	2.5888	3.0622	189.2297
61.2327	6.0	12	2.5881	3.0484	189.1622
61.2327	7.0	14	2.5876	3.0953	189.2027
61.2327	8.0	16	2.5865	3.1647	189.2838
61.2327	9.0	18	2.5870	2.9212	189.2162
61.2327	10.0	20	2.5870	3.1943	189.1284
61.2327	11.0	22	2.5858	3.0398	189.1959
61.2327	12.0	24	2.5851	3.1261	189.3514
61.2327	13.0	26	2.5840	3.0933	189.1622
61.2327	14.0	28	2.5827	3.1097	189.1081
61.2327	15.0	30	2.5814	2.8552	189.1081
61.2327	16.0	32	2.5779	2.889	189.0946
61.2327	17.0	34	2.5755	2.6436	189.0405
61.2327	18.0	36	2.5750	2.7995	188.8851
61.2327	19.0	38	2.5730	2.8456	189.0203
61.2327	20.0	40	2.5700	2.6827	189.2095
61.2327	21.0	42	2.5651	2.7145	188.9865
61.2327	22.0	44	2.5594	2.5647	188.8378
61.2327	23.0	46	2.5568	2.4516	188.75
61.2327	24.0	48	2.5550	2.5168	188.8176
61.2327	25.0	50	2.5475	2.4452	188.7365

Tsedeniya/finetuned_byT5large_kmr

finetuned_byT5large_kmr