learning_rate: 5e-05
train_batch_size: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
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
55.9442	2.0	88	3.3255	3.2534	188.4028
44.8760	4.0	176	2.6020	5.2143	181.2133
28.2708	6.0	264	1.6859	9.9555	182.9479
16.1472	8.0	352	1.1385	15.0211	223.4123
10.6747	10.0	440	0.9637	16.092	377.981
6.6775	12.0	528	0.7339	30.2746	107.4313
4.4425	14.0	616	0.7015	26.2359	378.6066
3.2021	16.0	704	0.6050	44.8457	174.1896
3.2021	18.0	792	0.6316	54.6453	105.3981
2.3825	20.0	880	0.5867	58.1638	107.3886
1.8789	22.0	968	0.6729	57.9448	99.2654
1.4530	24.0	1056	0.6139	60.0148	98.2654
1.0131	26.0	1144	0.6863	60.1916	99.6398
0.7077	28.0	1232	0.7421	61.4587	103.0758
0.5060	30.0	1320	0.6634	61.825	103.5972
0.4335	32.0	1408	0.7092	62.4758	113.8815
0.4335	34.0	1496	0.7753	60.204	88.1327
0.3493	36.0	1584	0.7520	62.0348	90.3175
0.3244	38.0	1672	0.7209	60.1887	103.3412

Tsedeniya/finetuned_byT5large_jje_joint

finetuned_byT5large_jje_joint