learning_rate: 0.0001
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 500
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Bleu	Em	F1
2.8703	1.0	2937	1.2938	0.5324	0.3045	0.489	0.2175	0.0387	0.5075
2.6182	2.0	5874	1.2576	0.5324	0.3067	0.49	0.2197	0.0433	0.5078
2.3196	3.0	8811	1.2556	0.5382	0.3136	0.4959	0.228	0.0431	0.5144
1.9782	4.0	11748	1.2703	0.5461	0.3219	0.5033	0.2368	0.0467	0.5226
1.9494	5.0	14685	1.2857	0.5456	0.3214	0.5036	0.2349	0.0477	0.5223
1.7425	6.0	17622	1.3035	0.5455	0.322	0.5025	0.2369	0.05	0.5222
1.6953	7.0	20559	1.3217	0.5461	0.322	0.5029	0.2365	0.0483	0.5226
1.5681	8.0	23496	1.3481	0.5458	0.3218	0.503	0.2381	0.05	0.5222
1.4350	9.0	26433	1.3630	0.5451	0.3214	0.5019	0.2368	0.0504	0.5218

Nigam/MCQ-QuestionGen

MCQ-QuestionGen