T5-QG2

This model is a fine-tuned version of Asif-Sheriff/T5-Question-Generation on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.9237

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAFACTOR and the args are: No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.3112	0.0548	100	1.7491
0.332	0.1095	200	1.7392
0.2815	0.1643	300	1.7695
0.298	0.2191	400	1.8053
0.2696	0.2738	500	1.8340
0.2625	0.3286	600	1.8321
0.2839	0.3834	700	1.8368
0.2651	0.4381	800	1.8531
0.2604	0.4929	900	1.8731
0.2798	0.5476	1000	1.8512
0.2148	0.6024	1100	1.8821
0.2603	0.6572	1200	1.8805
0.2092	0.7119	1300	1.9164
0.227	0.7667	1400	1.9262
0.2564	0.8215	1500	1.8991
0.2446	0.8762	1600	1.9290
0.2093	0.9310	1700	1.9291
0.243	0.9858	1800	1.9237

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1