VIT5-large-QA-Generation-checkpoint-2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5659

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: 5e-06
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• training_steps: 30000

Training results

Training Loss	Epoch	Step	Validation Loss
1.5918	1.8267	5000	1.6487
1.4528	3.6535	10000	1.6064
1.3326	5.4804	15000	1.5824
1.2677	7.3072	20000	1.5604
1.2456	9.1341	25000	1.5578
1.0773	10.9607	30000	1.5659

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0