learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
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: 8
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.6211	1.0	14778	1.4386	0.086	0.0686	0.0831	0.0831	7.0473
1.5116	2.0	29556	1.3540	0.0836	0.0677	0.0811	0.0811	6.9131
1.4459	3.0	44334	1.3019	0.0874	0.0708	0.0847	0.0846	7.1384
1.42	4.0	59112	1.2729	0.0843	0.0687	0.0818	0.0817	6.9433
1.3683	5.0	73890	1.2490	0.0838	0.0684	0.0814	0.0812	6.916
1.3589	6.0	88668	1.2357	0.0847	0.0692	0.0822	0.0821	6.995
1.353	7.0	103446	1.2245	0.0825	0.0673	0.08	0.0799	6.8302
1.3506	8.0	118224	1.2226	0.0823	0.0672	0.0799	0.0798	6.8086

gabbyhappy/MTSUSpring2025SoftwareEngineering

MTSUSpring2025SoftwareEngineering