araT5-Base-with-IA3

This model is a fine-tuned version of UBC-NLP/AraT5v2-base-1024 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5700
• Bleu: 5.2733
• Rouge: 0.2997
• Gen Len: 14.2112

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: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Rouge	Gen Len
8.4777	1.0	7500	4.1503	3.0695	0.2361	14.3284
5.3262	2.0	15000	3.8670	4.0626	0.2757	14.1688
5.0341	3.0	22500	3.7264	4.5492	0.2893	14.1428
4.879	4.0	30000	3.6486	4.945	0.2955	14.224
4.7908	5.0	37500	3.6025	5.1278	0.298	14.2372
4.7425	6.0	45000	3.5780	5.2779	0.2995	14.2132
4.7198	7.0	52500	3.5700	5.2733	0.2997	14.2112

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1